diff --git a/MNIST-classification.ipynb b/MNIST-classification.ipynb
index d8844bb..7c5a791 100644
--- a/MNIST-classification.ipynb
+++ b/MNIST-classification.ipynb
@@ -1,12 +1,71 @@
 {
  "cells": [
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Intro to machine learning. Classification\n",
+    "\n",
+    "Good articles to read: \n",
+    "\n",
+    "[MachineLearningForBeginners](https://towardsdatascience.com/introduction-to-machine-learning-for-beginners-eed6024fdb08)\n",
+    "\n",
+    "[AnotherMLForBeginners](https://towardsdatascience.com/machine-learning-for-beginners-d247a9420dab)\n",
+    "\n",
+    "[GreatMachineLearningCource](https://www.coursera.org/learn/machine-learning)\n",
+    "\n",
+    "\n",
+    "In general, \"machine learning is a field of study that gives the ability to the computer for self-learn without being explicitly programmed\" -  [Arthur Samuel](http://www.contrib.andrew.cmu.edu/~mndarwis/ML.html)\n",
+    "\n",
+    "Broadly speaking, machine learning can be divided into two classes: supervised and unsupervised learning. Within the field of machine learning, there are two main types of tasks: supervised, and unsupervised. The main difference between the two types is that supervised learning is done using a ground truth, or in other words, we have prior knowledge of what the output values for our samples should be. Therefore, the goal of supervised learning is to learn a function that, given a sample of data and desired outputs, best approximates the relationship between input and output observable in the data. Unsupervised learning, on the other hand, does not have labeled outputs, so its goal is to infer the natural structure present within a set of data points.Unsupervised learning tries to derive data patterns without explicitly specifying rules. \n",
+    "\n",
+    "In the following workshop we will be concentrating on supervised learning. One of the popular machine learning tasks is classification. Classification is an area of machine learning that gives computers the ability to automatically classify data. For example, some classification algorithms can recognize pictures and separate dogs from cats, another algorithm can detect spam vs not-spam from email content. In this exercise we will recognize hand written digits.\n"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "In this notebook, you'll learn how to use scikit-learn to do machine learning classification on the MNIST database of handwritten digits. \n",
+    "\n",
+    "![title](images/mnist-initial.png)\n",
+    "\n",
+    "\n",
+    "The MNIST dataset is a well-known dataset consisting of 28x28 grayscale images. For each image, we know the corresponding digits (from 0 to 9). It is available here: http://yann.lecun.com/exdb/mnist/index.html\n",
+    "While this tutorial uses a classifier called Logistic Regression, the coding process in this tutorial applies to other classifiers in sklearn (Decision Tree, K-Nearest Neighbors etc). In this noteboook, we use Logistic Regression to predict digit labels based on images. The image above shows a bunch of training digits (observations) from the MNIST dataset whose category membership is known (labels 0–9). After training a model with logistic regression, it can be used to predict an image label (labels 0–9) given an image. First lets load the data using mnist package"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### scikit-learn\n",
+    "\n",
+    "What is scikit-learn?\n",
+    "\n",
+    "Scikit-learn provides a range of supervised and unsupervised learning algorithms via a consistent interface in Python. The library is built upon the SciPy (Scientific Python) that must be installed before you can use scikit-learn. This stack that includes:\n",
+    "\n",
+    "NumPy: Base n-dimensional array package\n",
+    "SciPy: Fundamental library for scientific computing\n",
+    "Matplotlib: Comprehensive 2D/3D plotting\n",
+    "IPython: Enhanced interactive console\n",
+    "Sympy: Symbolic mathematics\n",
+    "Pandas: Data structures and analysis\n",
+    "Extensions or modules for SciPy care conventionally named SciKits. As such, the module provides learning algorithms and is named scikit-learn."
+   ]
+  },
   {
    "cell_type": "code",
-   "execution_count": 3,
+   "execution_count": 59,
    "metadata": {},
    "outputs": [],
    "source": [
+    "# This is a bit of magic to make matplotlib figures appear inline in the notebook\n",
+    "# rather than in a new window.\n",
+    "\n",
     "%matplotlib inline\n",
+    "# Some more magic so that the notebook will reload external python modules;\n",
+    "\n",
     "%reload_ext autoreload\n",
     "%autoreload 2"
    ]
@@ -15,38 +74,27 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### Download mnist set\n"
+    "Lets download the dataset first"
    ]
   },
   {
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "In this notebook, you'll learn how to use scikit-learn to do machine learning classification on the MNIST database of handwritten digits. \n",
-    "\n",
-    "![title](mnist-initial.png)\n",
-    "\n",
-    "\n",
-    "The MNIST dataset is a well-known dataset consisting of 28x28 grayscale images. For each image, we know the corresponding digits (from 0 to 9). It is available here: http://yann.lecun.com/exdb/mnist/index.html\n",
-    "While this tutorial uses a classifier called Logistic Regression, the coding process in this tutorial applies to other classifiers in sklearn (Decision Tree, K-Nearest Neighbors etc). In this noteboook, we use Logistic Regression to predict digit labels based on images. The image above shows a bunch of training digits (observations) from the MNIST dataset whose category membership is known (labels 0–9). After training a model with logistic regression, it can be used to predict an image label (labels 0–9) given an image. First lets load the data using mnist package"
+    "### Download mnist set"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": 5,
    "metadata": {},
    "outputs": [
     {
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "Requirement already satisfied: mnist in /home/ec2-user/anaconda3/envs/pytorch_p36/lib/python3.6/site-packages (0.2.2)\n",
-      "Requirement already satisfied: numpy in /home/ec2-user/anaconda3/envs/pytorch_p36/lib/python3.6/site-packages (from mnist) (1.15.4)\n",
-      "\u001b[31mfastai 1.0.40 requires dataclasses, which is not installed.\u001b[0m\n",
-      "\u001b[31mfastai 1.0.40 requires nvidia-ml-py3, which is not installed.\u001b[0m\n",
-      "\u001b[31mthinc 6.12.1 has requirement msgpack<0.6.0,>=0.5.6, but you'll have msgpack 0.6.0 which is incompatible.\u001b[0m\n",
-      "\u001b[33mYou are using pip version 10.0.1, however version 19.0.3 is available.\n",
-      "You should consider upgrading via the 'pip install --upgrade pip' command.\u001b[0m\n"
+      "Requirement already satisfied: mnist in /home/ubuntu/anaconda3/envs/python3_rl/lib/python3.6/site-packages (0.2.2)\r\n",
+      "Requirement already satisfied: numpy in /home/ubuntu/anaconda3/envs/python3_rl/lib/python3.6/site-packages (from mnist) (1.15.4)\r\n"
      ]
     }
    ],
@@ -69,12 +117,12 @@
     "\n",
     "SKLearn: Machine Learning framework\n",
     "\n",
-    "Seaborn: Creating viz"
+    "Seaborn: Creating vizualizations"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 6,
+   "execution_count": null,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -92,19 +140,38 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "## Explore data by splitting our data into train and test set"
+    "### Download datasets\n",
+    "\n",
+    "To automatically download the training images, training labels, test images and labels run the following cell.\n",
+    "\n"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 10,
+   "execution_count": 66,
    "metadata": {},
-   "outputs": [],
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/home/ec2-user/anaconda3/envs/pytorch_p36/lib/python3.6/site-packages/ipykernel/__main__.py:4: DeprecationWarning: `imresize` is deprecated!\n",
+      "`imresize` is deprecated in SciPy 1.0.0, and will be removed in 1.2.0.\n",
+      "Use ``skimage.transform.resize`` instead.\n",
+      "/home/ec2-user/anaconda3/envs/pytorch_p36/lib/python3.6/site-packages/ipykernel/__main__.py:4: DeprecationWarning: `toimage` is deprecated!\n",
+      "`toimage` is deprecated in SciPy 1.0.0, and will be removed in 1.2.0.\n",
+      "Use Pillow's ``Image.fromarray`` directly instead.\n"
+     ]
+    }
+   ],
    "source": [
     "\n",
+    "# get the train data\n",
     "train_images = mnist.train_images()\n",
+    "\n",
     "train_labels = mnist.train_labels()\n",
     "\n",
+    "# get test data\n",
     "test_images = mnist.test_images()\n",
     "test_labels = mnist.test_labels()\n",
     "\n",
@@ -139,7 +206,7 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "This proves in training we have 60000 images and each image is 28x28 dimension"
+    "This shows that in training dataset we have 60000 images and each image is 28x28 dimension"
    ]
   },
   {
@@ -151,27 +218,27 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 11,
+   "execution_count": 27,
    "metadata": {
     "scrolled": true
    },
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 576x576 with 9 Axes>"
       ]
      },
-     "metadata": {},
+     "metadata": {
+      "needs_background": "light"
+     },
      "output_type": "display_data"
     }
    ],
    "source": [
     "\n",
     "def print_digits(data_set, cols = 3, rows = 3):\n",
-    "    w=10\n",
-    "    h=10\n",
     "    fig=plt.figure(figsize=(8, 8))\n",
     "    for i in range(1, cols*rows +1):\n",
     "        img = data_set[i]\n",
@@ -185,7 +252,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 69,
+   "execution_count": 10,
    "metadata": {
     "scrolled": true
    },
@@ -266,24 +333,35 @@
     "### Preprocess data, normalization"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "So we have 60000 train_images brought in as a shape (28,28,60000) matrix. It is a numpy.ndarray. You need to convert it to an array of 1 dimensional images, meaning each image is represented as a single line/array of numbers, and you want 60000 arrays. In other words, you want to go from (28, 28, 60000) to (60000, 28*28).\n",
+    "\n",
+    "Lets use the reshape method for that.\n",
+    "\n"
+   ]
+  },
   {
    "cell_type": "code",
-   "execution_count": 12,
+   "execution_count": 73,
    "metadata": {},
    "outputs": [],
    "source": [
     "\n",
-    "x_train = train_images.reshape(-1,28*28)/255\n",
+    "x_train = train_images.reshape(train_images.shape[0],28,28)\n",
+    "\n",
     "y_train = train_labels\n",
     "\n",
-    "x_test = test_images.reshape(-1,28*28)/255\n",
+    "x_test = test_images.reshape(test_images.shape[0],28,28)\n",
     "y_test = test_labels\n",
     "\n"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 13,
+   "execution_count": 63,
    "metadata": {
     "scrolled": false
    },
@@ -292,12 +370,14 @@
      "name": "stdout",
      "output_type": "stream",
      "text": [
-      "10000 10000\n"
+      "length of the training data 60000 60000\n",
+      "length of the test data 10000 10000\n"
      ]
     }
    ],
    "source": [
-    "print(len(x_test),len(y_test))"
+    "print('length of the training data', len(x_train),len(y_train))\n",
+    "print('length of the test data', len(x_test),len(y_test))"
    ]
   },
   {
@@ -308,16 +388,14 @@
    "source": [
     "### Building Logistic regression with sklearn\n",
     "\n",
-    "Step 1. Import the model you want to use\n",
-    "\n",
-    "In sklearn, all machine learning models are implemented as Python classes\n",
+    "Now that we have preprocessed the data, lets go ahead and run some computation on it.\n",
     "\n",
-    "Over here we are importing LogisticRegression"
+    "Step1 -> import the ML algorithm you want to use. In sklearn, all machine learning models are implemented as Python classes .Over here we are importing LogisticRegression"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 15,
+   "execution_count": 16,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -329,9 +407,17 @@
     "\n"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Step2 - > The next line defines the logistic regression classifier. Since we have multiple outputs(10, for each digit), we need to use multi class classifier. Ove versus Rest is the simplest multilabel classifier. \n",
+    "\n"
+   ]
+  },
   {
    "cell_type": "code",
-   "execution_count": 16,
+   "execution_count": 17,
    "metadata": {},
    "outputs": [],
    "source": [
@@ -346,14 +432,14 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "Step 3. Training the model on the data, storing the information learned from the data\n",
+    "Step 3 -> Training the model on the data, storing the information learned from the data\n",
     "\n",
     "Model is learning the relationship between digits and labels"
    ]
   },
   {
    "cell_type": "code",
-   "execution_count": 21,
+   "execution_count": 18,
    "metadata": {},
    "outputs": [
     {
@@ -367,12 +453,12 @@
      "data": {
       "text/plain": [
        "LogisticRegression(C=1.0, class_weight=None, dual=False, fit_intercept=True,\n",
-       "          intercept_scaling=1, max_iter=100, multi_class='ovr', n_jobs=1,\n",
-       "          penalty='l2', random_state=None, solver='liblinear', tol=0.1,\n",
-       "          verbose=1, warm_start=False)"
+       "          intercept_scaling=1, max_iter=100, multi_class='ovr',\n",
+       "          n_jobs=None, penalty='l2', random_state=None, solver='liblinear',\n",
+       "          tol=0.1, verbose=1, warm_start=False)"
       ]
      },
-     "execution_count": 21,
+     "execution_count": 18,
      "metadata": {},
      "output_type": "execute_result"
     }
@@ -393,7 +479,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 22,
+   "execution_count": 19,
    "metadata": {},
    "outputs": [
     {
@@ -415,7 +501,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 23,
+   "execution_count": 20,
    "metadata": {
     "scrolled": true
    },
@@ -448,17 +534,21 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 41,
-   "metadata": {},
+   "execution_count": 21,
+   "metadata": {
+    "scrolled": true
+   },
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "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\n",
       "text/plain": [
        "<Figure size 648x648 with 2 Axes>"
       ]
      },
-     "metadata": {},
+     "metadata": {
+      "needs_background": "light"
+     },
      "output_type": "display_data"
     }
    ],
@@ -482,45 +572,72 @@
     "\n"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "If we wanted to see which images the model couldn't predict correctly:"
+   ]
+  },
   {
    "cell_type": "code",
-   "execution_count": 56,
+   "execution_count": 36,
    "metadata": {},
    "outputs": [
     {
      "data": {
-      "image/png": "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\n",
+      "image/png": "iVBORw0KGgoAAAANSUhEUgAABHsAAARwCAYAAAB6llgtAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDMuMC4yLCBodHRwOi8vbWF0cGxvdGxpYi5vcmcvOIA7rQAAIABJREFUeJzs3XmcXGWd7/HvjywsSYCw72kNmyxjGCKME5SOILLcARSQxUEY1ORelgGBKxkymkQcEUZkGRGNgtGRRQyERUHAkA4Gr0CiGImYYQtrSEwCYQuEpH/3j+dUUqlUn1Nbd5166vN+verV6Xqees6vT3d90/Xrc06ZuwsAAAAAAABx2KDZBQAAAAAAAKBxaPYAAAAAAABEhGYPAAAAAABARGj2AAAAAAAARIRmDwAAAAAAQERo9gAAAAAAAESkrZs9ZjbRzLzo9oqZ3WZmw3t5u1PNrKukjiVVPH5g8pgRDazpbDPzOtcYZGYvJftynzrXmpmsc0gNj234/imzjdOT+gZX+biOkp+5wu2W3qoV+UcWrbMmWVTdNmrKoqLHf8bMHjOzFWa21Mx+bWaDGl0nWgNZtM6aNWeRmX3SzB42s+VmtsjMppnZHnXWE2UWmdneSe68YmbvmdkLZvYjM9u+t2pF/pFF66xZTxbtmOTPW2a2xMy+a2ab1FlPrFkU5Wu0tm72JJZL+mhyu1DSCEnT+/iX3R9J+lQV8wdKmqBQa56Ml9S/3kXMbEdJH0s+PbmGJfK6f4pdqLU/dx+V9O/NLQc5QBY1DllUATP7oqSbJN0r6QhJX5T0lBqw79DSyKI6mNn+kn4l6WVJJ0g6U9IHJT1gZpvWuGbMWbSZpOcUftY+pVDnoZLuMTOyqL2RRXVInj/3SRom6URJ5ypk0uQ61ow5iwqieo1GiEqr3P33yb9/b2YvSPqtpCMl/aJ0spn1k9TP3Vc2qgB3f0nSS41arxnMbFdJ/6rwBLmuzuVOSj4+KOk4Mzuzkfs7J+YX/dwBElnUEGRRZcxsK0lXSjrH3X9YNDStSSUhP8ii+pwgaZmkU9x9lSSZ2VOS/iRplEJztVrRZpG7/07S74ru6jKzlyTdL+nvJP2hKYUhD8ii+pwg6UOSdnX35yTJzN6XdIuZTXL3p2pYM9osKhLVazSO7FnfnORjhySZ2RQzm21mx5rZPEnvSjowGdvFzG4xs2Vm9o6Z3Vd6mK6Z7Wxm9ySHyC9I/pKqkjnrHSJoZlua2Q/MbKGZvWtm883svGT4zeTjj4sOMSvUu5GZXW5mLyaHw/7JzI4sWXvD5DC+15Par5Q0oI59JklXKXS//1rnOlLoFP9e0mWSNpd0eOmEWvaPmXVamdM6zKzLzKYWff5RM7srOWT0bTN73Mw+14CvC6gGWVQbsqgyn00+/qRB6yFeZFF1Bkh6p9DoSbxe2FSNa8acReUsTT4O7MVtoPWQRdU5QtJjhUZP4g5JK1UmQyrUblnU8jiyZ30dycdXS+67XNLXJS2S9JyZbSFplsJ/SP9b0juSxkn6jZnt7u4rzMwk3SlpK0lfUAihSZK2UDhUviwz21hSl6Rtkvl/lbRrcpOkTyh0VL+hcKiwJC1MPk6VdIDCIXLPKPxCf5eZjXT3x5M531I4XH+8pL9I+pJC97e0jgWSutz99J5qTeYdKekfJJ0q6cNpc7OY2W6S9lc41HC6pMUKwXJX0Zxa909HhWUMk/SwpO8rfM9GKYRSt7vfnFJ7lyS5e2cF2/hx8jO0WNLNksa7+4oK60N76Eg+kkVkUW9k0YGS5kv6gpmNl7Stwl/Qv5z8pR0o6Eg+kkWVZdHPJJ1lZhcpnC4xWNJ3kpqnpzyurDbIosLcDRRel3xA4fvxmKRHK6wP7aEj+UgWVZZFeyZrrOHuK83smWSsKu2SRYrtNZq7t+1N0kRJSxT+c+kvaXdJMyS9IWn7ZM4USS5pRMljL1EIkS2K7huqcH7pWcnnRyaPPbBozjBJqxSeoOvUUfT5WEndpdssGh+crHt6yf2HJPcfXHL/Q5J+kfx7S0krJF1UNL6BwpPRSx73tKTrM/bhQIVQPDP5vDOpYZ8avydfk7S6aP9fK+ltSYMasH/K1qYQSlN7WMuSn40fSHqw6P7Tk7UGF903XdL0jK9ve0nflXR0Us/E5PtxZ7OfD9yadyOL1oyTRX2XRfcp/IXtZUmfU/jr3IPJz9y2zX5OcGvOjSxaM15zFiXzDpX0WrJtl/SkpF1q/J5EnUVFc39dtL9mS9qm2c8Hbs27kUVrxuv5vegpSVeVuX+WpJtq+J5EnUWK9DUap3GFJ9b7yW2+wkX0TnT3hUVzXva1HdeCQyU9IOkNM+tv4SJYbyocYjgymXOApEXu/kjhQe7+vNYehtiTT0j6Y5ltZjlUodv9cKGmpK7pRTXtK2kjhW52oabu4s+L7t/V3b+Qsc3zFTqrP6iy1p6cJGlm0f6/WdImCk+8glr3T0XMbKiZXWNmz2vtz8YYhf9oeuTuh7h76pXp3X2hu5/t7ne5e5e7T1TYh0dbL16ZHi2BLCKL1tGbWaTwC+RgSV9w9xvd/deSjlX4Re7s+qtHCyOL6sgiM9tb4cLntyfbP0ah8XOP1XaB5tizqOAcrT0yc7Cke81sozrKRusji+r7vUgKTY9S1sP9WaLOolhfo3EaV+jyHqrwQ/+qpFc8ae8VWVTmcVsp/Kd0YpmxwmG62ykcAlZqsaQhKTVtqbWH/FVjq2Sb75cZW11UU6GG0pqqYmZbKxxmeLqkIeGISBXe5m6ImQ1y97erWG+EwoXEfmRmmyd3z5P0isJhgoXD82rdP5WaovC9vUTh8Mc3JP0fhV/YesNUSd+T9PeSeiUc0RLIovKfZyKLarIs+dhVuMPd3zCzOZL2asD6aF1kUfnPK3WJpKeKX4iZ2W8VLvL6RYVTuirSJlkkSfK1F4t9JNlfz0k6RdINjdoGWg5ZVP7zSr2mcF2dUptr7XXEKtJOWVSi5V+j0ewJV3qfnTGnXPdzmcI5ipeUGStcfOpVhXMWS22jcFhYT5Zq7bmN1VimcEj+sSlzCue5bqO1v+wXPq/WjgovqKaWGfudQqAeWsV6hbfwuyK5FdvazIa6+2uqff+8m3wsveDfFgqHiir5K9JRks529+8XJiTnkvcWL/mI9kQWrf28WmRR9Z5U+HkqvWCsKRyCjfZFFq39vBZ7Kpxusoa7v5b8JXp4lWu1Qxatx92fN7NlCkdyoH2RRWs/r8VfVXJtHjMbqPC8+n7ZR/SsLbNIEbxG4zSu2k2XtLekee4+u+Q2P5nzmKRtzezAwoPMbBeF7mDW2vuZ2d/1MF54i7vSw1unK3SF3ypTUyEs/6zwhFrTAU2eJLV0RJ+WNLrk9uVk7AxJF1S6UHKhtBMVfkEqXfMUhSvRH1f0ddayfwpvnfihou3uLKn46vwbSuon6b2iOUO07iGKjXZ88jHr0FGgHLKILKrFLxUaO6OL1t9M4eKLf2rQNtBeyKLgeUn7Fd9hZlsqXIB0QaWLtFEWrcfCuyZtqXB0D1Atsii4V9JHzGxY0X1HKzynf13pIu2cRYrhNZrn4MJBzbqp5KJbPcyZIml2mfu3kvSCpP+n8IN+sMJV1a+VdHIyxxQO+XpBoSP6aYUn8stKv/jXRgq/bL+qcBX50QovWL5VNOdZhYtpHqRwrufAZHu/kvSiwjUXRisExARJlxY99mqFMLlA4aKctyk8ybzka6zoQoQlj+lU+QtsTZG0IOVxo5LHndjD+FwlF9aqdf8k9z+ajH1aIZjmKBxuOLXosY8q/IJxXDLvkeQxxd+j01Xbxb8mKnTEP6NwpMHXFf6CcFuznw/cmncji8iivs6iZN4dyTZPU/hr2UxJf5M0tNnPCW7NuZFF9WdRsr4n++lTCv/fP6Jw2sQOJftxQco6bZFFkr6t8A5En07qPlOhKfa0ii78yq29bqUZ0MOcKSKL0vbPAElPJM/pI5Ov81VJPyuzHxekrNMuWTRREb5Ga3oBTf3i6wiSZGwHST9WOF/0PYX/nH4mae+iObsodE9XKPy1Z6zCqQZdaXUo/EXjhwrnab6rcCjevxaNH5Y8ud5NfqA7kvs3VHiru6cVuqevJts/quixGyqcf7hc4XzO/1K4AJWX1LBA0pQq92mnyr/AulXSoymP+25Sz8Y9jH9F614Bvtb9s6vCNSreVrjY2zEqudJ7MufBZM4LybbX+R71ECRdxd/XHr6OkxTeZWJ58v15OgmTDZv9fODWvBtZRBb1dRYl8wZLuk7hsOsVkn4jad9mPx+4Ne9GFjUmixReWD6mcD2JxZLu0frvGEQW+Zrfix5WOG3lnaTuKyRt1eznA7fm3cplQJk5U0QWZe3HnRT+sPOWwv/110rapGQOWeTxvkaz5IsDelVyrvrX3P0nza4FQPsiiwDkAVkEIA/IorjR7EGvM7MdFDqlHe6+Mms+APQGsghAHpBFAPKALIofzR4AAAAAAICI8G5cAAAAAAAAEaHZAwAAAAAAEBGaPQAAAAAAABHpX8+DzexwSVdL6ifpR+7+rYz5XCAIaCHubs2uoRJkERC3Vskiqbo8IouAlrPE3bdudhGVIIuAqFWURTUf2WNm/SRdK+kISXtJOtnM9qp1PQCoBVkEIC/IIyB6zze7gEqQRUD0Ksqiek7jOkDS0+7+bPJWbbdIOqaO9QCgFmQRgLwgjwDkAVkEoK5mz46SXiz6/KXkPgDoS2QRgLwgjwDkAVkEoK5r9pQ7f3698z3NbIykMXVsBwDSkEUA8iIzj8giAH2ALAJQV7PnJUk7F32+k6RXSie5+2RJkyUu/gWgV5BFAPIiM4/IIgB9gCwCUNdpXI9J2s3MPmBmAyWdJOmuxpQFABUjiwDkBXkEIA/IIgC1H9nj7qvM7GxJ9ym8pd8N7j6vYZUBQAXIIgB5QR4ByAOyCIAkmXvfHbXHIYJAa3H3cud8tzyyCGgtZBGAnJjj7iObXUSjkUVAy6koi+o5jQsAAAAAAAA5Q7MHAAAAAAAgIjR7AAAAAAAAIkKzBwAAAAAAICI0ewAAAAAAACJCswcAAAAAACAiNHsAAAAAAAAiQrMHAAAAAAAgIjR7AAAAAAAAIkKzBwAAAAAAICI0ewAAAAAAACJCswcAAAAAACAiNHsAAAAAAAAiQrMHAAAAAAAgIjR7AAAAAAAAIkKzBwAAAAAAICI0ewAAAAAAACJCswcAAAAAACAiNHsAAAAAAAAiQrMHAAAAAAAgIjR7AAAAAAAAIkKzBwAAAAAAICI0ewAAAAAAACJCswcAAAAAACAiNHsAAAAAAAAiQrMHAAAAAAAgIjR7AAAAAAAAIkKzBwAAAAAAICI0ewAAAAAAACJCswcAAAAAACAiNHsAAAAAAAAiQrMHAAAAAAAgIv2bXUA7GzBgQOr4/vvvn7nGKaeckjo+c+bMzDV22mmn1PF99903dfyMM87I3IaZZc7pCytWrEgdf/jhhzPXOPDAA1PHv/nNb2aucfnll6eOd3d3Z64BAAD63iabbJI6fvTRR2eu8fGPfzx1/JhjjslcY4cddkgdf+KJJzLXWLJkSer46NGjM9cAAOQTR/YAAAAAAABEpK4je8xsgaQ3Ja2WtMrdRzaiKACoFnkEIA/IIgB5QBYBaMRpXKPdPf0YUADoG+QRgDwgiwDkAVkEtDFO4wIAAAAAAIhIvc0el3S/mc0xszHlJpjZGDObbWaz69wWAKRJzSOyCEAfIYsA5AFZBLS5ek/jGuXur5jZNpIeMLO/uvtDxRPcfbKkyZJkZl7n9gCgJ6l5RBYB6CNkEYA8IIuANlfXkT3u/krycbGkaZIOaERRAFAt8ghAHpBFAPKALAJQc7PHzAaZ2ZDCvyUdJumJRhUGAJUijwDkAVkEIA/IIgCSZO61HbVnZh9U6BJL4XSwm9z9PzIe0zaHCHZ0dGTOmTRpUur4qaee2qBq6rN69erU8Up+hswsdbxfv35V1ZRnWftLkjbddNPU8RUrVjSqnLq4e/o3LieqzaN2yqJW0tnZWdd4V1dX5jYqmYP8IYtQqYEDB6aO77DDDplr/PM//3Pq+AknnJA6vs8++2RuoxH++Mc/po4vX748c43//M//TB3/9a9/XVVNbWBOK7yFOVkUh5Ej03/U9t1339TxD37wg40sp1cNHz48dfyZZ56pexuvv/565pz77rsvdXzrrbfOXGPmzJmp493d3ZlrVKCiLKr5mj3u/qykD9f6eABoFPIIQB6QRQDygCwCIPHW6wAAAAAAAFGh2QMAAAAAABARmj0AAAAAAAARodkDAAAAAAAQEZo9AAAAAAAAEaHZAwAAAAAAEJGa33q93W244Yap4zNmzMhcY9iwYanjS5cuzVxj3LhxqeMHH3xw5hovvfRS6vhdd92VOr5q1arMbWywQXpfcfjw4ZlrtIrnn38+c86KFSv6oBIgPzo7OzPnVJKb9erq6ur1bQBY3+DBgzPnnHPOOanjQ4cOzVzjrLPOSh3P+n1EkgYMGJA6vnLlytTxX/7yl5nbuOOOO1LHf/7zn2eu8f7776eOH3nkkZlrnHTSSanjY8aMyVwja86SJUsy1wCwvrvvvjt1fJtttumjSupjZplz3L3X66jkNevll19e93YeffTR1PGf/exnmWt8//vfTx1fvXp1RbVwZA8AAAAAAEBEaPYAAAAAAABEhGYPAAAAAABARGj2AAAAAAAARIRmDwAAAAAAQERo9gAAAAAAAESEZg8AAAAAAEBEaPYAAAAAAABEpH+zC2hV/fun77phw4ZlrvHaa6+lju+7776Za7z66qup49dff33mGnnw6KOPNrsEAHWYOHFi6viECRP6phAAuXTRRRdlzvnMZz6TOr7nnnvWXceSJUsy59x7772p49/97ndTx2fPnl1VTc106qmn1r3GV77yldTxSvY5gPWdcsopqePnnXder9cwd+7czDlvvfVW6vizzz6buYa7V1xTrRYsWJA5p6OjI3X8E5/4ROYaO+64Y+r4pZdemrnGcccdV3cdEkf2AAAAAAAARIVmDwAAAAAAQERo9gAAAAAAAESEZg8AAAAAAEBEaPYAAAAAAABEhGYPAAAAAABARGj2AAAAAAAARKR/swtoZ6tXr04df/XVV/uoEgDtbMaMGanjnZ2dfVJHV1dX6vikSZPqejyA5nn00Ucz55x77rl1b+eb3/xm6vi3v/3tzDVWrlyZOr5ixYqqamqWrbfeOnPOO++8kzr+zDPPZK6xbNmyimsCULms38+yxlG92bNnp45PnTq17m1svvnmmXM22KAxx+RwZA8AAAAAAEBEaPYAAAAAAABEhGYPAAAAAABARGj2AAAAAAAARIRmDwAAAAAAQERo9gAAAAAAAESEZg8AAAAAAEBEaPYAAAAAAABEpH+zC2hVu+22W+r4Cy+8kLnGxhtvnDq+ySabZK7xzjvvZM4B0L7cvdklSJJGjx6dOaerq6v3CwHQFHfffXfmnNNOOy11/LDDDstc49JLL00dj+n3pq9+9aup42eddVbmGv37p78UqGSNZcuWZc4BYtKvX7/MORMmTEgd33TTTeteY/ny5ZlrIH9ef/31PttW5pE9ZnaDmS02syeK7tvCzB4ws6eSj0N7t0wAII8A5ANZBCAPyCIAaSo5jWuKpMNL7hsnabq77yZpevI5APS2KSKPADTfFJFFAJpvisgiAD3IbPa4+0OSSo/PPEbST5J//0TSsQ2uCwDWQx4ByAOyCEAekEUA0tR6zZ5t3X2hJLn7QjPbpqeJZjZG0pgatwMAWSrKI7IIQC8jiwDkAVkEQFIfXKDZ3SdLmixJZpaPK4UCaDtkEYA8IIsA5AFZBMSv1rdeX2Rm20tS8nFx40oCgKqQRwDygCwCkAdkEQBJtTd77pJUeH/M0yTd2ZhyAKBq5BGAPCCLAOQBWQRAkmTu6UftmdnNkjolbSVpkaQJku6QdKukXSS9IOkEdy+9OFi5tdrmEMFttunxMkZrzJ07N3X8sssuy1zjyiuvrLgmoFrubs2uoVij8iimLJoxY0bqeGdnZ5/UYZarHxVEhixCK9lwww0z52y88cap4+eff37mGhdffHHq+OrVqzPXOOOMM1LHb7zxxsw12swcdx/Z7CIKyKLmOOywwzLn3HvvvXVv5/jjj08dnzZtWt3bQMuqKIsyr9nj7if3MHRI1SUBQB3IIwB5QBYByAOyCECaWk/jAgAAAAAAQA7R7AEAAAAAAIgIzR4AAAAAAICI0OwBAAAAAACICM0eAAAAAACAiNDsAQAAAAAAiAjNHgAAAAAAgIj0b3YBsVq2bFnmnB/84Aep45dddlnmGgMGDEgdv/zyyzPXAJBPEydOzJzT2dnZ63WMHj2617eBdVXyfc2aU8nPD4Dq7bDDDqnjRxxxROYakydPrruOuXPnpo6fc845mWvMmjWr7jqA2Gy22Wap4+PHj++TOmbOnNkn20G8OLIHAAAAAAAgIjR7AAAAAAAAIkKzBwAAAAAAICI0ewAAAAAAACJCswcAAAAAACAiNHsAAAAAAAAiQrMHAAAAAAAgIv2bXUCsVq1alTlnxYoVqeP9+2d/ewYOHFhxTQBQzqRJk1LHu7q6+qaQPtDZ2VnXuCRNmDChMcX0skrqzPrejh49ukHVAK2jX79+qeNjxoxJHT/22GMzt7F48eLU8VNOOSVzjblz56aOL126NHMNAOv7l3/5l9Txgw46KHMNd08dnzp1auYay5cvz5wDpOHIHgAAAAAAgIjQ7AEAAAAAAIgIzR4AAAAAAICI0OwBAAAAAACICM0eAAAAAACAiNDsAQAAAAAAiAjNHgAAAAAAgIiYu/fdxsz6bmMt4B//8R9Tx3/1q19lrrHBBun9uqOOOipzjVmzZmXOQXtyd2t2Db2hVbKor/LZrDW+zZ2dnanjEyZMqHsNrKurqytzzujRo3u9DrIIjTJ8+PDMOWPHjk0dv+CCC1LHv/GNb2RuY9KkSanj3d3dmWugKea4+8hmF9FoZNG6HnroodTxUaNGZa5xxx13pI4fd9xxVdUElKgoiziyBwAAAAAAICI0ewAAAAAAACJCswcAAAAAACAiNHsAAAAAAAAiQrMHAAAAAAAgIjR7AAAAAAAAIkKzBwAAAAAAICI0ewAAAAAAACLSv9kFtLPf/e53qeNdXV2ZaxxzzDGp45/85Ccz15g1a1bmHAB9r5IM6OzsrHs7WWtUUkcjZNUxY8aMPqkDazXi5wvIk0suuSRzzoknnljXNsysrscDaK6PfexjqePunrnGokWLGlUOULPMI3vM7AYzW2xmTxTdN9HMXjazx5Pbkb1bJoB2RxYByAvyCEAekEUA0lRyGtcUSYeXuf9Kdx+R3O5pbFkAsJ4pIosA5MMUkUcAmm+KyCIAPchs9rj7Q5KW9UEtANAjsghAXpBHAPKALAKQpp4LNJ9tZnOTwweH9jTJzMaY2Wwzm13HtgCgJ2QRgLzIzCOyCEAfIIsA1NzsuU7ScEkjJC2UdEVPE919sruPdPeRNW4LAHpCFgHIi4ryiCwC0MvIIgCSamz2uPsid1/t7t2SfijpgMaWBQDZyCIAeUEeAcgDsghAQU3NHjPbvujTT0t6oqe5ANBbyCIAeUEeAcgDsghAQf+sCWZ2s6ROSVuZ2UuSJkjqNLMRklzSAklje7FG1GHWrFl1r/HhD384dfzZZ5/NXGPixImp40uWLMlcY4899kgdf/DBBzPXmD9/fur4I488krkGmqMds2jmzJmZczo7O+vezoQJE3p9G5U4+OCD+2Q7qNzo0aObXUIutWMexeKBBx7InLPvvvumju+www6p4+PHj8/cxt577506/vnPfz5zjbfffjtzDuJGFvWO5cuXp44PGTIkc4399tsvdXzHHXfMXOPll1/OnAOkyWz2uPvJZe6+vhdqAYAekUUA8oI8ApAHZBGANPW8GxcAAAAAAAByhmYPAAAAAABARGj2AAAAAAAARIRmDwAAAAAAQERo9gAAAAAAAESEZg8AAAAAAEBEaPYAAAAAAABExNy97zZm1ncbi8Ctt96aOef4449PHX/uuecy11i5cmXq+C677JI6/sADD2Ru45BDDkkdf+qppzLX2HvvvVPHzSxzjayf93HjxmWucfXVV6eOd3d3Z67RKtw9e6e2oFbJos7Ozsw5M2bM6P1CkEuTJk2qe42urq66xvsKWYQ82X333VPHK8nu733ve6njV111VeYaF154YeYcNNwcdx/Z7CIajSxaV9brgYkTJ2auMWDAgNTx1157LXON+fPnp46/9957qeOVvJb87//+79Txt99+O3MNNEVFWcSRPQAAAAAAABGh2QMAAAAAABARmj0AAAAAAAARodkDAAAAAAAQEZo9AAAAAAAAEaHZAwAAAAAAEBGaPQAAAAAAABExd++7jZn13cYisP3222fOOffcc1PH33zzzcw1tt1229TxAw88MHX8c5/7XOY2uru7U8ffeOONzDWGDRuWOr7xxhtnrvGpT30qdXz8+PGZa1xwwQWp41deeWXmGq3C3a3ZNfSGmLKos7MzdXzGjBl9UwjW6OrqypwzadKkutdoJ2QRYvPee++ljlfy+/m4ceNSx6+66qqqakJF5rj7yGYX0WhkUXV+/OMfZ875/Oc/3weV1G/p0qWp42PHjs1cY9q0aY0qB5WrKIs4sgcAAAAAACAiNHsAAAAAAAAiQrMHAAAAAAAgIjR7AAAAAAAAIkKzBwAAAAAAICI0ewAAAAAAACJCswcAAAAAACAiNHsAAAAAAAAiYu7edxsz67uNoWIDBgxIHd9gg/Se4HvvvdfIcnrVkCFDUsfnzJmTucaWW26ZOr7//vtnrrFgwYLMOXng7tbsGnoDWbSuzs7OusaxrokTJza7hOiQRYjN/PnzU8d33XXXzDUuvvji1PHLLrusqppQkTnuPrLZRTQaWVSd2267LXPOsccemzp+4YUXZq7x7rvvpo5vuummqePjxo3L3EbWGi+++GLmGh0dHZlz0HAVZRFH9gAAAAAAAESEZg8AAAAAAEBEaPYAAAAAAABEhGYPAAAAAABARGj2AAAAAAAARIRmDwAAAAAAQERo9gAAAAAAAESkf7MLQPNAWeCvAAAgAElEQVS9//77zS6hzwwYMCB1fLPNNstcY+jQoanjHR0dmWssWLAgcw7QV7q6uuoaB4B+/fqljn/sYx/LXOOxxx5LHX/77berqinPFi9enDq+66679lElAKr1l7/8JXPOsccemzp+xhlnZK5x0EEHpY4vX748dXzUqFGZ2zjqqKMy56B1ZR7ZY2Y7m9kMM3vSzOaZ2bnJ/VuY2QNm9lTyMf0VMADUgSwCkAdkEYA8IIsAZKnkNK5Vki5w9w9J+gdJZ5nZXpLGSZru7rtJmp58DgC9hSwCkAdkEYA8IIsApMps9rj7Qnf/Q/LvNyU9KWlHScdI+kky7SeS0o9VA4A6kEUA8oAsApAHZBGALFVdoNnMOiTtJ+kRSdu6+0IphI2kbRpdHACUQxYByAOyCEAekEUAyqn4As1mNljSbZLOc/c3zKzSx42RNKa28gBgXWQRgDwgiwDkAVkEoCcVHdljZgMUQuRGd789uXuRmW2fjG8vqezbCrj7ZHcf6e4jG1EwgPZFFgHIA7IIQB6QRQDSVPJuXCbpeklPuvt3iobuknRa8u/TJN3Z+PIAICCLAOQBWQQgD8giAFkqOY1rlKRTJf3ZzB5P7rtY0rck3WpmX5D0gqQTeqdEAJBEFgHIB7IIQB6QRQBSZTZ73H2WpJ5O/jykseUAPdt8881Tx4866qjMNa655prU8aFDh2ausWDBgtTxZ599NnMNVI8sApAHZFF52223Xer41KlTM9e4//77U8dPOeWUqmrqLZtuumnq+GmnnZY6Lkkf/ehHG1UO2hRZ1Dx33pl9sNTYsWNTx/faa6/MNe6+++7U8cmTJ6eOf+QjH8ncBuJW1btxAQAAAAAAIN9o9gAAAAAAAESEZg8AAAAAAEBEaPYAAAAAAABEhGYPAAAAAABARGj2AAAAAAAARIRmDwAAAAAAQET6N7sAtIcjjjgidXzPPffMXOPMM89MHR8+fHjmGu+++27q+KWXXpq5xsyZM1PHX3jhhcw1AACIycsvv5w6fvLJJ2euMWfOnNTxQw45JHONgw46KHV8l112yVxjn332SR3fbrvtUsd32mmnzG385S9/SR3fY489MtdYuXJl5hwAjTd79uzMOYcddljq+I033pi5xqhRo+oab4S7776717eB3sORPQAAAAAAABGh2QMAAAAAABARmj0AAAAAAAARodkDAAAAAAAQEZo9AAAAAAAAEaHZAwAAAAAAEBGaPQAAAAAAABGh2QMAAAAAABARc/e+25hZ320sAh0dHZlzNt1009Txww8/PHONnXbaqdKSytpyyy0z55x00kmp42aWuUZ3d3fq+Lx58zLXuP7661PHr7nmmsw12om7Z39jWhBZBLQWsqg9DB06NHX88ccfz1yj3t9p+krW7yO/+c1vMte49dZb665jwIABqeNZv2dK0tKlS+uuo4XMcfeRzS6i0ciivjdo0KDMOeeff37q+ODBg1PHTzzxxMxtXH311anj1157beYaK1euzJyDhqsoiziyBwAAAAAAICI0ewAAAAAAACJCswcAAAAAACAiNHsAAAAAAAAiQrMHAAAAAAAgIjR7AAAAAAAAIkKzBwAAAAAAICLm7n23MbO+21gEdt9998w5Z511Vur4xhtvnLnG5ptvnjp+/PHHp44//PDDmdvYc889U8fnzZuXucbPf/7z1PHrrrsucw1Ux92t2TX0BrIIaC1kUXsYMGBA6vh9992XucbWW2+dOr7XXntlrtHd3Z06Pnny5NTx22+/PXMbDz74YOp4X/1+3r9//9TxQYMGZa6xfPnyRpXTCua4+8hmF9FoZBHQcirKIo7sAQAAAAAAiAjNHgAAAAAAgIjQ7AEAAAAAAIgIzR4AAAAAAICI0OwBAAAAAACICM0eAAAAAACAiNDsAQAAAAAAiAjNHgAAAAAAgIiYu6dPMNtZ0k8lbSepW9Jkd7/azCZK+pKkvyVTL3b3ezLWSt8YgFxxd2t2DQVkEdC+yCIAOTHH3Uc2uwiJLALaXEVZVEmzZ3tJ27v7H8xsiKQ5ko6V9FlJb7n7tyutiCABWkvOXmCRRUCbIosA5ESemj1kEdC+Ksqi/lkT3H2hpIXJv980sycl7Vh/fQBQObIIQB6QRQDygCwCkKWqa/aYWYek/SQ9ktx1tpnNNbMbzGxog2sDgLLIIgB5QBYByAOyCEA5FTd7zGywpNsknefub0i6TtJwSSMUuspX9PC4MWY228xmN6BeAG2OLAKQB2QRgDwgiwD0JPOaPZJkZgMk/VLSfe7+nTLjHZJ+6e77ZKzD+aBAC8nTdTIksghoV2QRgJzIzTV7JLIIaGMVZVHmkT1mZpKul/RkcYgkFwUr+LSkJ2qpEgAqQRYByAOyCEAekEUAsmReoFnSKEmnSvqzmT2e3HexpJPNbIQkl7RA0theqRAAArIIQB6QRQDygCwCkKqi07gatjEOEQRaSt5OnWgUsghoLWQRgJzI1WlcjUIWAS2nMadxAQAAAAAAoHXQ7AEAAAAAAIgIzR4AAAAAAICI0OwBAAAAAACICM0eAAAAAACAiNDsAQAAAAAAiAjNHgAAAAAAgIjQ7AEAAAAAAIgIzR4AAAAAAICI0OwBAAAAAACICM0eAAAAAACAiNDsAQAAAAAAiAjNHgAAAAAAgIjQ7AEAAAAAAIgIzR4AAAAAAICI0OwBAAAAAACISP8+3t4SSc8Xfb5Vcl/etUqdUuvUSp2N1Rt1DmvwennSqlkktU6t1NlY7VwnWZRPrVIrdTZWu9cZax6VZpHE97rRqLOxWqVOqYm/G5m7N3i7lTOz2e4+smkFVKhV6pRap1bqbKxWqTOvWmn/tUqt1NlY1NkeWmn/tUqt1NlY1Nk+WmUfUmdjUWfjNbNWTuMCAAAAAACICM0eAAAAAACAiDS72TO5yduvVKvUKbVOrdTZWK1SZ1610v5rlVqps7Gosz200v5rlVqps7Gos320yj6kzsaizsZrWq1NvWYPAAAAAAAAGqvZR/YAAAAAAACggWj2AAAAAAAARKRpzR4zO9zM5pvZ02Y2rll1ZDGzBWb2ZzN73MxmN7ueAjO7wcwWm9kTRfdtYWYPmNlTycehzayxoIdaJ5rZy8l+fdzMjmxyjTub2Qwze9LM5pnZucn9udunKbXmap+2CrKofq2SR62QRUlNLZFHZFFjkUX1I4saiyxqT2RR/ciixiKL6qipGdfsMbN+kv5H0iclvSTpMUknu/tf+ryYDGa2QNJId1/S7FqKmdnHJb0l6afuvk9y3+WSlrn7t5JwHuruFzWzzqSucrVOlPSWu3+7mbUVmNn2krZ39z+Y2RBJcyQdK+l05WyfptT6WeVon7YCsqgxWiWPWiGLpNbJI7KocciixiCLGossaj9kUWOQRY1FFtWuWUf2HCDpaXd/1t1XSrpF0jFNqqUluftDkpaV3H2MpJ8k//6Jwg9X0/VQa664+0J3/0Py7zclPSlpR+Vwn6bUiuqRRQ3QKnnUClkktU4ekUUNRRY1AFnUWGRRWyKLGoAsaiyyqHbNavbsKOnFos9fUn5D2SXdb2ZzzGxMs4vJsK27L5TCD5ukbZpcT5azzWxucghh0w9lLDCzDkn7SXpEOd+nJbVKOd2nOUYW9Z5cP3dK5PZ50yp5RBbVjSzqPbl93pSR2+cNWdQ2yKLek9vnTRm5fd6QRdVpVrPHytyX1/eAH+Xufy/pCElnJYe7oX7XSRouaYSkhZKuaG45gZkNlnSbpPPc/Y1m15OmTK253Kc5RxYht8+bVskjsqghyCLk9nlDFrUVsgi5fd6QRdVrVrPnJUk7F32+k6RXmlRLKnd/Jfm4WNI0hcMb82pRcq5g4ZzBxU2up0fuvsjdV7t7t6QfKgf71cwGKDwxb3T325O7c7lPy9Wax33aAsii3pPL506pvD5vWiWPyKKGIYt6T+6eN+Xk9XlDFrUdsqj35O55U05enzdkUW2a1ex5TNJuZvYBMxso6SRJdzWplh6Z2aDk4koys0GSDpP0RPqjmuouSacl/z5N0p1NrCVV4YmZ+LSavF/NzCRdL+lJd/9O0VDu9mlPteZtn7YIsqj35O65U04enzetkkdkUUORRb0nV8+bnuTxeUMWtSWyqPfk6nnTkzw+b8iiOmryJrwblyRZeMuxqyT1k3SDu/9HUwpJYWYfVOgUS1J/STflpU4zu1lSp6StJC2SNEHSHZJulbSLpBckneDuTb/oVg+1diocyuaSFkgaWzjnshnM7CBJv5X0Z0ndyd0XK5xnmat9mlLrycrRPm0VZFH9WiWPWiGLpNbJI7Koscii+pFFjUUWtSeyqH5kUWORRXXU1KxmDwAAAAAAABqvWadxAQAAAAAAoBfQ7AEAAAAAAIgIzR4AAAAAAICI0OwBAAAAAACICM0eAAAAAACAiNDsAQAAAAAAiAjNHgAAAAAAgIjQ7AEAAAAAAIgIzR4AAAAAAICI0OwBAAAAAACICM0eAAAAAACAiNDsAQAAAAAAiAjNHgAAAAAAgIi0dbPHzCaamRfdXjGz28xseC9vd6qZdZXUsaSKxw9MHjOigTWdbWZe42N3NLNpZvaWmS0xs++a2SZ11jMz+Z4cUsNjG75/ymzj9KS+wTU8dm8zu9/M3kn213W1rIN4kEXrrFlPFn3SzB42s+VmtijJpT3qrCfaLCpaYwMzm5Os878aWR9aC1m0zpo1ZZGZdZbsw8LtW3XWE20WmdnI5PeipWa2zMx+Y2YH9kadaA1k0Tpr1ppFB5vZDDNbbGbvmdmzZnaFmW1aZz1RZpGZdfSQ3bf0Vq19oa2bPYnlkj6a3C6UNELSdDMb1Ic1/EjSp6qYP1DSBIVam8rM+ku6T9IwSSdKOlfSCZIm17HmjpI+lnx6cg1L5Gb/lDKzzSQ9KGljhf11oaTjJP2smXUhF8iiOpjZ/pJ+JellhQw6U9IHJT1Q6y82MWdRiS9K2rHZRSA3yKLG+JzW7sePSrq21oViziIz21nSbyT1l/R5Sacm/77fzIY1szY0HVlUny0k/VHSWQpfwxWSTpN0U60LxpxFRS7Uutn9780tpz79m11ADqxy998n//69mb0g6beSjpT0i9LJZtZPUj93X9moAtz9JUkvNWq9PnaCpA9J2tXdn5MkM3tf0i1mNsndn6phzZOSjw9KOs7Mzmzk/m6yMxUaPf/k7q9Lkpktk3SnmY1099lNrQ7NRBbV5wRJyySd4u6rJMnMnpL0J0mjJN1bw5oxZ5EkycyGSvoPSeMUfqkFyKLGmOvuTzRorZiz6ChJQyR9puj3ot9JWqLwM3ddE2tDc5FFdXD3aZKmFd3VZWYrJU02sy3cfVkNy8acRQXzi37uWh5H9qxvTvKxQ5LMbIqZzTazY81snqR3JR2YjO1iZrckh5y+Y2b3WckpA2a2s5ndY2YrzGyBmX2xdIPlDhE0sy3N7AdmttDM3jWz+WZ2XjL8ZvLxx0WHmBXq3cjMLjezF5ND9v5kZkeWrL2hhVOtXk9qv1LSgBr31xGSHis0ehJ3SFop6fAa1zxZ0u8lXSZp83Lr1LJ/bO2h1fuUrNVlZlOLPv+omd2VHDL6tpk9bmafq/FrKTVC0uzCLzSJ+yW5wi88QAFZVJ0Bkt4pNHoSheeZ1bhmzFlUcImkhyVNb/C6iAdZ1HwxZ9EASaskvVV031vJfbVmN+JEFtVvafJxYI2PjzmLosSRPevrSD6+WnLf5ZK+LmmRpOfMbAtJsxSeNP9b0jsKfxn9jZnt7u4rzMwk3SlpK0lfUAihSQqH1fV4xIuZbSypS9I2yfy/Sto1uUnSJxQ6qt9QOG1BkhYmH6dKOkDhELlnJH1W0l0Wjhp5PJnzLYXD9sdL+oukLyn8Vby0jgWSutz99J5qlbRnssYa7r7SzJ5JxqpiZrtJ2l/hdLDpkhYrBMtdRXNq3T8dFZYxTOHFz/cVvmejFEKp291vTqm9S5LcvTNl7Y0UGmHFVknqVjhCCijoSD6SRZVl0c8knWVmFymcRjpY0neSmqtuZLRBFsnM/k7Sv0j6cIX1oD11JB/JosqyqOBBM9tS0osKR81d6u6rK3hc6TZjz6LbFH6OrjCz/0ju+5qk11Tm6A20tY7kI1lURRZZOOKpv8LrjH+XdLu7v5r+qLLrxJ5FBT9OfoYWS7pZ0nh3X1Fhffnj7m17kzRR4TDR/sltd0kzJL0haftkzhSFoy5GlDz2EoUQ2aLovqEK55eelXx+ZPLYA4vmDFN4cd9VWkfR52MVXvyP6KHuwcm6p5fcf0hy/8El9z8k6RfJv7eUtELSRUXjGyg8Gb3kcU9Luj5jHz4l6aoy98+SdFMN35OvSVpdtP+vlfS2pEEN2D+dyf37lNzfJWlqD2tZ8rPxA0kPFt1/erLW4KL7pkuanvH1XaEQagOK7jswWev+Zj8nuDXnRhatGa85i5J5hyq8QPDk9qSkXWr8nkSdRcm8mZIuT/7dkazzv5r9fODWvBtZtGa8nt+L9pN0afK1HirpqiRLrq7xe9IOWTRC4VSZQna/IunDzX4+cGvejSxaM17X70XJ3L8WPbd+LWmTGr8nUWeRpO0lfVfS0Uk9E5Pvx53Nfj7Uc+M0rvDEej+5zVe4oOeJ7r6waM7LvrbjWnCopAckvWFm/S1cqPhNhUMMRyZzDpC0yN0fKTzI3Z/X2sMQe/IJSX8ss80shyp0ux8u1JTUNb2opn0Vji65s6im7uLPi+7f1d2/UMF2vcx91sP9WU6SNLNo/98saROFJ15BrfunImY21MyuMbPntfZnY4zCfzQ9cvdD3D3ryvQ/lLS1pP8ys+3MbG9J31MIz6r/4oeokEV1ZFHyXLpJ0u3J9o9RaPzcY7VdoDnqLDKzkyTtofDXNaAYWVRHFrn7H93939z9Hnf/jbufJ+nbks40s62qrF+KP4u2VzjiYY7CpQGOSP79KzPbpRH1o2WRRfW/RpPCG8GMUjjKaV9Jv0iObKpW1Fnk7gvd/Wx3v8vdu9x9oqTzJR1tvfjuYb2N07hCl/dQhcbEq5Je8aS9V2RRmcdtJekfFN5RqVThlIHtFA4BK7VY4WJ0PdlSaw/5q8ZWyTbfLzNWaCRsV1RDaU21eE3hnM1Sm2vt9TIqkjyRPiTpR2ZWWHOewl94TlYIFan2/VOpKQrf20sUDqF8Q9L/UXjxWBd3/6uZjZF0pdZ2vycr/PyV+zlD+yCLyn9eqUskPVX8y4+Z/Vbhr8VfVDilqyKxZ5GZDZD0nwrn3G+QfI2FhtggMxvi7m/2uABiRxaV/7weUyV9RdLfKZzCUJHYsyjxfxVejxzv7u9Lkpk9qHDk+IWS/rUB20BrIovKf14Vd5+X/PN3ZvakwlG9o0UWVWKqwh/l/15SrzSwehvNnnCl96x3QCp3hMoyhXMULykzVvgl+VWFcxZLbaNwWFhPlmrtuY3VWKbwtsPHpswpnKO5TTK/uKZa/FUl1+Yxs4EK3ffvV7lW4S38rkhuxbY2s6Hu/ppq3z/vJh9LL0q2hcKhojKzjRQulHy2u6+p38wadhScu99gZjdJ2k0hwJcofE28E057I4vWfl6LPRUO8V7D3V9L/vozvMq1Ys+iQZJ2UmiAlTbBblG4lkAtXxfiQBat/bzRqj3iOfYskkJ2zys0eqQ1136cp+qzG3Ehi9Z+3ih/SD5+UFU0e9QeWVSOl3xsOZzGVbvpkvZW+A9qdsltfjLnMUnbmtmBhQclh6T+fQVr75dcPLOcwgV+NyrzuO0kvVWmpkJY/lnhCbWmA5o8SWrtiN4r6SNmNqzovqMlbahwXmhFksMJT1R4sTa65HaKwpXoj0um17p/Cm+duOZCyGa2s8KpDAUbSuon6b2iOUO07iGKdXP3d939z+6+SNI/KzwXb23kNtA2yKLgeYVrZayRXBy1Q9KCShdpkyx6S+t/bYVf5C6WxDtboBZkUc+OU7gWyNxKH9AmWSSF7N4n+UNhYf0NJe2jKrIbKEIW9WxU8vG5Sh/QRllUzvHJx6zT+3KLI3tq9x2FF+kPmtl/KXRrt5V0sKRZHq4Ifo+kPymcG3mRwhP468o+HO+nks6SdL+ZTVQ4T/UDknZ393HJXzyek/RZM3siWXeuwvmp90l6wMwuUzi8blOFC99tlJxDvtTMJkuaZGarkjlfUrhg1jrM7GmFczPTzgmdqnDF+NvN7KuSNlM4Rekmd19zNXszmyKp0907eljnHxUujHaRu3eVqeXfFF6M/KjW/ePuL5nZY5IuMbN3FBosF6uoe+7uy5M5XzOzNxROsxqncChp6nU/zGx6skaP54Qm1w4Zr3BBtlUKQXmBpC+5+7KeHgekIIuC70u6I8mamxWOXrlI4ReLG4vWmqI2zyIPb0/fVfKYwv74c/E1DIAqkEVhznWS/qbwYnKlwoVgz1Z4M4ulRfOmqM2zKPEjhVNtp5nZ9xSu+XiWwsVSJ6etD/SALApz/lvS/yicfvSOQiPrK5L+n4qOhCaL1syZqHAK38MKp4d9XOE009vdveJGfe54Dq4S3aybSq6w3sOcKZJm9zC2g6QfK5wv+p7CXyB+Jmnvojm7KBzhskLhrxdjFRokXWl1KJzz+EOF0HlX4XSpfy0aP0whPN5VOLSsI7l/Q4W3unta4ZeMV5PtH1X02A0Vzj9crnDNnf9SuACVl9SwQNKUCvbjTpLuUPhr8VKFq7NvUjLnVkmPpqzx3aSejXsY/4rWvQJ8rftnV4UXOW8rBNAxKrnSezLnwWTOC8m21/keqfyV3ruKv689fB2DJN2vEF4rFH4ZPLbZzwVuzb2RRQ3Los8mz6k3knrv0frv0kEWlf+6OsS7cbX9jSyqP4sUrjEzV+F0kfcUXrCdJ2mDknlk0dp5hyj8EWxZcpup8OKz6c8Jbs25lcuAMnOmiCxK2z/nKByRslzhNdqfJX21+DmazCOLwpyTJM1OvtaVyffp65I2bPbzoZ6bJV8c0KssXDfja+7+k2bXAqB9kUUA8oAsApAHZFHcaPag15nZDgqd0g53X5k1HwB6A1kEIA/IIgB5QBbFj2YPAAAAAABARHg3LgAAAAAAgIjQ7AEAAAAAAIhIn771uplxzhjQQtzdml1DbyCLgNZCFgHIiSXuvnWzi2g0sghoORVlUV1H9pjZ4WY238yeNrNx9awFALUiiwDkBXkERO35ZhdQKbIIiFpFWVRzs8fM+km6VtIRkvaSdLKZ7VXregBQC7IIQF6QRwDygCwCINV3ZM8Bkp5292eTt2q7RdIxjSkLACpGFgHIC/IIQB6QRQDqavbsKOnFos9fSu5bh5mNMbPZZja7jm0BQE/IIgB5kZlHZBGAPkAWAajrAs3lLpa43sW93H2ypMkSF/8C0CvIIgB5kZlHZBGAPkAWAajryJ6XJO1c9PlOkl6prxwAqBpZBCAvyCMAeUAWAair2fOYpN3M7ANmNlDSSZLuakxZAFAxsghAXpBHAPKALAJQ+2lc7r7KzM6WdJ+kfpJucPd5DasMACpAFgHIC/IIQB6QRQAkydz77hRNzgcFWou7lzvnu+WRRUBrIYsA5MQcdx/Z7CIajSwCWk5FWVTPaVwAAAAAAADIGZo9AAAAAAAAEaHZAwAAAAAAEBGaPQAAAAAAABGh2QMAAAAAABARmj0AAAAAAAARodkDAAAAAAAQEZo9AAAAAAAAEaHZAwAAAAAAEBGaPQAAAAAAABGh2QMAAAAAABARmj0AAAAAAAARodkDAAAAAAAQEZo9AAAAAAAAEaHZAwAAAAAAEBGaPQAAAAAAABGh2QMAAAAAABARmj0AAAAAAAARodkDAAAAAAAQEZo9AAAAAAAAEaHZAwAAAAAAEBGaPQAAAAAAABGh2QMAAAAAABARmj0AAAAAAAAR6d/sAgAAAIBGGD9+fOr4JZdckrnGk08+mTre2dmZOv63v/0tcxtZ9txzz8w5S5cu7fU6AOTXG2+8kTo+ZMiQzDW6u7vrrmPatGmp4xMnTsxc44knnqi7DqyPI3sAAAAAAAAiQrMHAAAAAAAgIjR7AAAAAAAAIkKzBwAAAAAAICI0ewAAAAAAACJCswcAAAAAACAiNHsAAAAAAAAiYu7edxsz67uNoWGOOuqo1PFPf/rTmWvsscceqeO33HJL5hrXXntt5hw0lrtbs2voDWRRnA466KDMOb/97W9Tx7/85S9nrnHVVVdVXBMagyyCVNnvGz/96U9TxzfZZJPMNbJ+N77mmmtSx88///zMbYwfPz51fNy4cZlrHHHEEanjs2bNylwDVZvj7iObXUSjkUX5NGjQoNTxhQsX1vV4KTvvGuGVV17JnDNt2rTU8RtuuCFzjXnz5qWOr1q1KnONFlJRFvWvZwtmtkDSm5JWS1oVY/gBaA3kEYA8IIsA5AFZBKCuZk9itLsvacA6AFAv8ghAHpBFAPKALALaGNfsAQAAAAAAiEi9zR6XdL+ZzTGzMY0oCABqRB4ByAOyCEAekEVAm6v3NK5R7v6KmW0j6QEz+6u7P1Q8IQkXAgZAb0vNI7IIQB8hiwDkAVkEtLm6juxx91eSj4slTZN0QJk5k919JBcFA9CbsvKILALQF8giAHlAFgGoudljZoPMbEjh35IOk/REowoDgEqRRwDygCwCkAdkEQCpvtO4tpU0zcwK69zk7r9uSFUAUB3yCEAekEUA8oAsAiBz977bmFnfbaxN9O+f3q/7p3/6p8w1vv71r6eO77PPPqnjq1atytzGypUrU8cHDhyYucYNN9yQOj527NjMNVAdd7dm19AbyKLWtNlmm6WOz5o1K3ONvffeO3V84cKFmWuMGDEidfxvf/tb5hpbb7116viJJ56YucaUKVNSx996663MNVoFWQRJquR31u7u7tTx5MVvqqzn8MEHH5w6/qEPfShzG7fddlvq+Kmnnpq5xo033pg5Bw03J8bTnsiifDrnnHNSx6+88srU8UryLut13HPP/X/27jerN9YAACAASURBVD1MrqpKFPjaJsAQkMvDGEAdEBEYRC+PKIjCoCioM0IQUBF5jGIQxRnFi/hkmFERYZRBnBGDYgLDyHUIAe4gL3kpKoOJgoSngEGQCAlvCSCYff+oijah++zuruquU6d/v++rr7tr7V57pbprpXv1Oad+Xcyx/vrrV8bXXHPNYo5u+PjHP14ZP+mkk8aljnEyrF7kpdcBAAAAGsSwBwAAAKBBDHsAAAAAGsSwBwAAAKBBDHsAAAAAGsSwBwAAAKBBDHsAAAAAGmRyrwtgaJtssklxzZe+9KXK+Dvf+c6O65g7d25l/LjjjivmmD9/fmX8G9/4RjHHcB4PoLmmT59eGd9yyy073uPBBx8srlmyZEnH+5x22mmV8be97W3FHKU1e+21VzHHU089VVwDdbF8+fLimpxzZXzevHnFHEcccURlfMqUKZXx008/vbjHzTffXBkfTp1A/1pvvfWKa/bdd98xr+P++++vjG+xxRbFHDvssENl/LWvfW0xx/77718Z33zzzYs5ttpqq+KaicaRPQAAAAANYtgDAAAA0CCGPQAAAAANYtgDAAAA0CCGPQAAAAANYtgDAAAA0CCGPQAAAAANYtgDAAAA0CCTe13ARLbVVltVxr/2ta8Vc7zhDW+ojN97773FHPvuu29l/Kc//WllPOdc3KPkjjvuKK456KCDKuNvfetbizkuvPDCYdcEjK/NNtusMv7Vr351zGv44Ac/2HGOj33sY8U1b3rTmzre55WvfGVlfKONNirmuO222zquA8bL855X/hvl8uXLK+P77LNPx3W85S1vqYxPmTKlmGPvvfeujC9btmxENQH9ZY899iiu2XHHHce8jnvuuafjHNdcc01H8YiIE088sTI+nMfi0UcfLa6ZaBzZAwAAANAghj0AAAAADWLYAwAAANAghj0AAAAADWLYAwAAANAghj0AAAAADWLYAwAAANAgk3tdQFOtuuqqxTXf+ta3KuPbb799McdPfvKTyvhBBx1UzHH77bcX14y13/zmN8U1q6++emX8ta99bTHHhRdeWBnfbbfdijnuv//+yvh1111XzAE813777VcZ32qrrTre47zzzquM//SnPy3mOPnkkyvjhxxySDFH6f+IJUuWFHN84QtfqIzfdtttxRzQT+bOnVtcM2PGjMr4XnvtVcxx8803V8bnzJlTGb/pppuKe9xyyy3FNQBj7cwzz+x1CcNS+p2XwTmyBwAAAKBBDHsAAAAAGsSwBwAAAKBBDHsAAAAAGsSwBwAAAKBBDHsAAAAAGsSwBwAAAKBBDHsAAAAAGmRyaUFK6bSI+NuIuD/nvFX7vnUj4v9GxMYRsSgi3plzfmjsyuw/Rx55ZHHN9ttvXxm/5557ijkOPfTQyvjtt99ezFEHZ599dnHNWWedVRk/88wzizmmT59eGT/33HOLOebNm1cZ33///Ys5GB39qH/ttttuxTWf/exnO9rjqaeeKq55//vfXxnfY489ijkOPvjgyviqq65azPHMM89Uxo866qhijjlz5hTXMDb0ot4YTo/YfffdK+Pvfe97O65jypQplfGjjz664z1gOPQioMpwjuyZHRFvWem+T0bEZTnnl0fEZe2PAcba7NCPgN6bHXoR0HuzQy8ChlAc9uScfxgRD650954RseJPinMiYkaX6wJ4Dv0IqAO9CKgDvQioMtpr9kzLOS+OiGi/fWH3SgIYEf0IqAO9CKgDvQiIiGFcs6dTKaWZETFzrPcBqKIXAXWgFwF1oBdB8432yJ77UkobRES0394/1MKc86yc8/Scc/WVcQFGZ1j9SC8CxpheBNSBXgRExOiHPedHxEHt9w+KiPO6Uw7AiOlHQB3oRUAd6EVARAxj2JNS+m5E/DQiNk8p3ZNSen9EHBcRb04p/Soi3tz+GGBM6UdAHehFQB3oRUCV4jV7cs77DRHatcu1NMr73ve+jnPMmjWruGbhwoUd71MHy5cvL675yEc+Uhn//e9/X8wxd+7cyvjqq69ezHHhhRcW1zA29KP6Wm211Srju+++ezHH85432oNNW84+++zimlVWWaUyftRRRxVzTJkyZdg1DeWCCy6ojM+ZM6cyTm/pRb1xyy23FNfMmzevMr7//vsXc+ScK+O/+MUvOqoBukUvosqyZcuKa84444xxqIRe6ewnawAAAABqxbAHAAAAoEEMewAAAAAaxLAHAAAAoEEMewAAAAAaxLAHAAAAoEEMewAAAAAaxLAHAAAAoEEm97qAfrXGGmtUxlddddWO93j44Yc7ztEkS5curYyvv/76xRy77bZbZfyBBx4o5rjqqquKa2Ci+cxnPlMZ/+hHP9rxHvfcc09l/O/+7u+KOa677rrK+JZbbjmimgZz+eWXF9cccMABHe8DPNeBBx5YGV+2bFkxxwc+8IHK+HbbbVcZP/vss4t7HHbYYZXxJUuWFHMA/evmm28urvnhD39YGd95550r46uvvnpxj3e84x2V8e985zvFHNSXI3sAAAAAGsSwBwAAAKBBDHsAAAAAGsSwBwAAAKBBDHsAAAAAGsSwBwAAAKBBDHsAAAAAGmRyrwvoV695zWsq4y9+8YuLOe65557K+BlnnDGimia6Pffcs7hm8uTqb/nhPOZ33333sGuCfrD66qtXxr/zne8Uc+yxxx7dKmdI66+/fmX8xhtvLObYbLPNKuM552KOp59+ujL+sY99rJjj8ccfL64BeuOmm26qjN96662V8RkzZhT32G677SrjRxxxRDHHvHnzimuAerrmmmuKa3bffffK+AMPPFAZX2ONNYp7fPnLX66Mr7baasUcp5xySnENveHIHgAAAIAGMewBAAAAaBDDHgAAAIAGMewBAAAAaBDDHgAAAIAGMewBAAAAaBDDHgAAAIAGmdzrAvrVIYcc0nGO6667rjL+8MMPd7xHk2y55ZaV8RNOOKHjPY499tiOc0C/mTlzZmV83333HadKqk2eXP1f1stf/vJxqeOCCy6ojC9cuHBc6gBGbueddy6uueqqqyrjRx99dGX8N7/5TXGPj370o5Xxs88+u5jjoIMOqoyfc845xRzLli0rrgF64+mnn66Mr7XWWpXxK664orhHqSd+4QtfKOa4+uqrK+N+LuodR/YAAAAANIhhDwAAAECDGPYAAAAANIhhDwAAAECDGPYAAAAANIhhDwAAAECDGPYAAAAANIhhDwAAAECDTO51Af1q3rx5lfH3vOc9xRwbb7xxZXzKlCnFHMuWLSuu6Qerr756cc1JJ51UGX/+85/fcR05545zQL/ZcccdK+MppXGqZOx149+y5ZZbVsb32GOPYo7zzz+/4zqA59piiy0q45tvvnkxxze/+c3K+JIlSyrjRxxxRHGPiy++uDI+Z86cYo7SmhkzZhRz7LPPPsU1QH+65JJLimt22mmnyvjaa69dzPGxj32sMv6hD32omOOpp54qrmHkikf2pJROSyndn1JaOOC+Y1JKv00pXde+vW1sywQmOr0IqAv9CKgDvQioMpzTuGZHxFsGuf/EnPPW7dv3u1sWwHPMDr0IqIfZoR8BvTc79CJgCMVhT875hxHx4DjUAjAkvQioC/0IqAO9CKjSyQWaD08p/bJ9+OA6Qy1KKc1MKc1PKc3vYC+AoehFQF0U+5FeBIwDvQgY9bDnGxHxsojYOiIWR8RXhlqYc56Vc56ec54+yr0AhqIXAXUxrH6kFwFjTC8CImKUw56c83055z/mnJdHxKkR8ZrulgVQphcBdaEfAXWgFwErjGrYk1LaYMCHe0XEwqHWAowVvQioC/0IqAO9CFgh5ZyrF6T03YjYJSJeEBH3RcQ/tj/eOiJyRCyKiENzzouLm6VUvVmD3HHHHcU1m2yySWV8v/32K+Y466yzhl1TL/2v//W/KuNz584t5th11127Vc6Qpk6dWlyzdOnSMa+jLnLOqdc1rKAXjc60adOKa26//fbK+JQpUzqu47HHHiuuWWONNSrjz3teJ5eZa0mp+lu69H/icFx88cXFNe973/sq47/73e86rqNJ6tSLIrrXjyZSLxovM2fOrIyfcsopxRzTp1ef1fLzn/98RDWNxhZbbFFc88Mf/rAyvt566xVzHH300ZXxL37xi8UcE8yCOp32pBdRZZVVVimu+e///u/KeDd+/zrggAOKa7773e92vM8EM6xeNLm0IOc82MTh26MqCWCU9CKgLvQjoA70IqBK538mBQAAAKA2DHsAAAAAGsSwBwAAAKBBDHsAAAAAGsSwBwAAAKBBDHsAAAAAGsSwBwAAAKBBJve6gKZasGBBcc0mm2xSGT/uuOOKOW644YbK+E033VTMkXOujE+aNKky/upXv7q4xxe+8IXK+Pbbb1/MMXfu3Mr43nvvXcwBE83b3/72jnMsW7asuOb444+vjH/3u9/tOMeee+5ZzFFy5513Vsbvv//+Yo5Sz9tll12KObbZZpvK+IUXXljMAYxc6WeeurjllluKa9761rdWxi+44IJijhkzZlTGv/jFLxZzAPW0fPny4pqlS5eOeR2lPhMxvJ8TGTlH9gAAAAA0iGEPAAAAQIMY9gAAAAA0iGEPAAAAQIMY9gAAAAA0iGEPAAAAQIMY9gAAAAA0SMo5j99mKY3fZj222mqrFdecddZZlfEZM2Z0XMd5551XXPP0009XxtdZZ53K+K677lrcY+HChZXxD3/4w8Ucq6++emX8oosuKuYomTp1anHN0qVLO96nX+ScU69rGAsTqRfVxb777ltcc+aZZ1bGJ02a1HEd3cjxhje8oTL+q1/9qpjjnnvu6biOiUQvYri22GKLyvhNN91UzHHEEUdUxv/1X/91RDX1ysyZM4trTjnllMp46bGI6J/Ho0sW5Jyn97qIbtOLmqnUDyPKv6N1w9/+7d8W13Tj97gJZli9yJE9AAAAAA1i2AMAAADQIIY9AAAAAA1i2AMAAADQIIY9AAAAAA1i2AMAAADQIIY9AAAAAA1i2AMAAADQICnnPH6bpTR+m/WB1VdfvTK+zz77FHN8+ctfroyvv/76xRwppcr4woULK+PnnHNOcY/jjz++Mv74448Xc+y+++6V8YsuuqiY49Zbb62Mb7vttsUcy5YtK65pipxz9TdHn9KLum/69OmV8R/96EfFHKuuumpHNfzsZz8rrtlhhx062oPe0IvolhtvvLG4ZsmSJZXxXXbZpUvVjK3Xv/71xTVXXXVVZbz0WEQM72fNBlmQc67+D68P6UX96WUve1ll/PLLLy/meNGLXtRxHY8++mhl/M1vfnMxx4IFCzquY4IZVi9yZA8AAABAgxj2AAAAADSIYQ8AAABAgxj2AAAAADSIYQ8AAABAgxj2AAAAADSIYQ8AAABAg0zudQET2RNPPFEZP+OMM4o5Smt23333Yo7Jk6u/DS688MLK+PLly4t7dMNGG23UcY6bb765Mr5s2bKO94CJaKuttqqMr7rqqh3v8dhjj1XG//7v/77jPYBmO+ecc4prPv3pT1fGSz97HXDAASOqaaw873nlv+mW1qSUulUOMELHHHNMZfz9739/ZXzDDTcs7pFzHklJg7rgggsq4wsWLOh4D0an+L9ASuklKaUrUko3p5RuTCn9Q/v+dVNKl6aUftV+u87YlwtMVHoRUAd6EVAHehFQMpzTuJ6JiI/nnP8qInaIiA+nlLaMiE9GxGU555dHxGXtjwHGil4E1IFeBNSBXgRUKg57cs6Lc84/b7//WETcHBEviog9I2JOe9mciJgxVkUC6EVAHehFQB3oRUDJiK7Zk1LaOCK2iYj/iYhpOefFEa1mk1J64RCfMzMiZnZWJsCf6UVAHehFQB3oRcBghj3sSSmtGRFzI+KjOedHh3vBtpzzrIiY1c7R+RWggAlNLwLqQC8C6kAvAoYyrJdeTymtEq0mcmbOecXLGNyXUtqgHd8gIu4fmxIBWvQioA70IqAO9CKgynBejStFxLcj4uac81cHhM6PiIPa7x8UEed1vzyAFr0IqAO9CKgDvQgoGc5pXK+LiAMi4oaU0nXt+z4dEcdFxPdSSu+PiN9ExL5jUyJAROhFQD3oRUAd6EVApeKwJ+d8dUQMdfLnrt0th267+OKLe13CsEyaNKm4Zq+99hqHSqgrvaje9t577zHf4/rrr6+MX3vttWNeA+hF/e1LX/pScc1f/dVfVcZnzKh+caNPfepTXamjZOrUqZXxQw45pJhj+fLllfFjjz12RDUxfvSiwe2xxx6V8RNOOKHjPc4888zimrvvvrsyfuihhxZzbLPNNpXx4fz+1Km77rqruOawww4b8zoYnWFdswcAAACA/mDYAwAAANAghj0AAAAADWLYAwAAANAghj0AAAAADWLYAwAAANAghj0AAAAADZJyzuO3WUrjtxl9Ze211y6ueeihhzre59xzz62M77XXXh3v0SQ559TrGsaCXtR9v/jFLyrjr3rVq4o57rzzzsr4DjvsUBl/4IEHinvQn/QixtMWW2xRGf/MZz5TGZ8xY0ZxjzXXXLMyPnfu3GKOLbfcsjK++eabF3NccskllfG3vvWtxRwTzIKc8/ReF9FtTepFW2+9dWV8/vz541RJ7z3zzDPFNSeccEJl/KSTTirmWLp06bBromuG1Ysc2QMAAADQIIY9AAAAAA1i2AMAAADQIIY9AAAAAA1i2AMAAADQIIY9AAAAAA1i2AMAAADQIIY9AAAAAA0yudcFAND/7rjjjsr4q171qmKO5cuXV8affvrpEdUEMBq33HJLZfyAAw6ojO+1117FPXbbbbeOc0ydOrUyftNNNxVzlP4t0G9++ctfVsY333zzYo7Pfe5zlfH3vve9I6ppMD/5yU+Ka1ZdddXK+OOPP14Z32OPPYp7lHLQ3xzZAwAAANAghj0AAAAADWLYAwAAANAghj0AAAAADWLYAwAAANAghj0AAAAADWLYAwAAANAgk3tdAERE7LnnnuOyz3nnnTcu+8BEc8cdd1TGFy1aVMxxzDHHVMYfffTREVQE0Bvz5s3reM1hhx3WrXJgQlm+fHllvPTzSkTEwQcf3FEc6sKRPQAAAAANYtgDAAAA0CCGPQAAAAANYtgDAAAA0CCGPQAAAAANYtgDAAAA0CCGPQAAAAANYtgDAAAA0CAp51y9IKWXRMTpEbF+RCyPiFk555NSSsdExAciYkl76adzzt8v5KreDKiVnHPqdQ0r6EUwcelFQE0syDlP73UREXoRTHDD6kXDGfZsEBEb5Jx/nlJ6fkQsiIgZEfHOiPh9zvlfhluRRgL9pWa/YOlFMEHpRUBN1GnYoxfBxDWsXjS5tCDnvDgiFrfffyyldHNEvKjz+gCGTy8C6kAvAupALwJKRnTNnpTSxhGxTUT8T/uuw1NKv0wpnZZSWmeIz5mZUpqfUprfUaUAbXoRUAd6EVAHehEwmOJpXH9amNKaEXFVRHwx53xOSmlaRCyNiBwRn4/WYYTvK+RwiCD0kTqdOrGCXgQTj14E1ERtTuNaQS+CCWlYvWhYR/aklFaJiLkRcWbO+ZyIiJzzfTnnP+acl0fEqRHxmk6qBSjRi4A60IuAOtCLgCrFYU9KKUXEtyPi5pzzVwfcv8GAZXtFxMLulwfQohcBdaAXAXWgFwElxQs0R8TrIuKAiLghpXRd+75PR8R+KaWto3WI4KKIOHRMKgRo0YuAOtCLgDrQi4BKw75mT1c2cz4o9JU6XiejG/Qi6C96EVATtbtmTzfoRdB3unfNHgAAAAD6g2EPAAAAQIMY9gAAAAA0iGEPAAAAQIMY9gAAAAA0iGEPAAAAQIMY9gAAAAA0iGEPAAAAQIMY9gAAAAA0iGEPAAAAQIMY9gAAAAA0iGEPAAAAQIMY9gAAAAA0iGEPAAAAQIMY9gAAAAA0iGEPAAAAQINMHuf9lkbEXQM+fkH7vrrrlzoj+qdWdXbXWNS5UZfz1Um/9qKI/qlVnd01kevUi+qpX2pVZ3dN9Dqb2o9W7kURvtbdps7u6pc6I3r4s1HKOXd53+FLKc3POU/vWQHD1C91RvRPrersrn6ps6766fHrl1rV2V3qnBj66fHrl1rV2V3qnDj65TFUZ3eps/t6WavTuAAAAAAaxLAHAAAAoEF6PeyZ1eP9h6tf6ozon1rV2V39Umdd9dPj1y+1qrO71Dkx9NPj1y+1qrO71Dlx9MtjqM7uUmf39azWnl6zBwAAAIDu6vWRPQAAAAB0Uc+GPSmlt6SUbk0p3Z5S+mSv6ihJKS1KKd2QUroupTS/1/WskFI6LaV0f0pp4YD71k0pXZpS+lX77Tq9rHGFIWo9JqX02/bjel1K6W09rvElKaUrUko3p5RuTCn9Q/v+2j2mFbXW6jHtF3pR5/qlH/VDL2rX1Bf9SC/qLr2oc3pRd+lFE5Ne1Dm9qLv0og5q6sVpXCmlSRFxW0S8OSLuiYifRcR+Oeebxr2YgpTSooiYnnNe2utaBkop7RwRv4+I03POW7XvOz4iHsw5H9duzuvknI/qZZ3tugar9ZiI+H3O+V96WdsKKaUNImKDnPPPU0rPj4gFETEjIg6Omj2mFbW+M2r0mPYDvag7+qUf9UMviuiffqQXdY9e1B16UXfpRROPXtQdelF36UWj16sje14TEbfnnO/MOf8hIs6KiD17VEtfyjn/MCIeXOnuPSNiTvv9OdH65uq5IWqtlZzz4pzzz9vvPxYRN0fEi6KGj2lFrYycXtQF/dKP+qEXRfRPP9KLukov6gK9qLv0oglJL+oCvai79KLR69Ww50URcfeAj++J+jblHBGXpJQWpJRm9rqYgmk558URrW+2iHhhj+spOTyl9Mv2IYQ9P5RxhZTSxhGxTUT8T9T8MV2p1oiaPqY1pheNnVo/d1ZS2+dNv/QjvahjetHYqe3zZhC1fd7oRROGXjR2avu8GURtnzd60cj0atiTBrmvri8L9rqc87YR8daI+HD7cDc6942IeFlEbB0RiyPiK70tpyWltGZEzI2Ij+acH+11PVUGqbWWj2nN6UXU9nnTL/1IL+oKvYjaPm/0oglFL6K2zxu9aOR6Ney5JyJeMuDjF0fEvT2qpVLO+d722/sjYl60Dm+sq/va5wquOGfw/h7XM6Sc83055z/mnJdHxKlRg8c1pbRKtJ6YZ+acz2nfXcvHdLBa6/iY9gG9aOzU8rmzsro+b/qlH+lFXaMXjZ3aPW8GU9fnjV404ehFY6d2z5vB1PV5oxeNTq+GPT+LiJenlF6aUlo1It4dEef3qJYhpZTWaF9cKVJKa0TEbhGxsPqzeur8iDio/f5BEXFeD2uptOKJ2bZX9PhxTSmliPh2RNycc/7qgFDtHtOhaq3bY9on9KKxU7vnzmDq+Lzpl36kF3WVXjR2avW8GUodnzd60YSkF42dWj1vhlLH541e1EFNuQevxhURkVovOfavETEpIk7LOX+xJ4VUSCltEq1JcUTE5Ij4z7rUmVL6bkTsEhEviIj7IuIfI+LciPheRPxlRPwmIvbNOff8oltD1LpLtA5lyxGxKCIOXXHOZS+klF4fET+KiBsiYnn77k9H6zzLWj2mFbXuFzV6TPuFXtS5fulH/dCLIvqnH+lF3aUXdU4v6i69aGLSizqnF3WXXtRBTb0a9gAAAADQfb06jQsAAACAMWDYAwAAANAghj0AAAAADWLYAwAAANAghj0AAAAADWLYAwAAANAghj0AAAAADWLYAwAAANAghj0AAAAADWLYAwAAANAghj0AAAAADWLYAwAAANAghj0AAAAADTKhhz0ppWNSSnnA7d6U0tyU0svGeN+zU0pXrlTH0hF8/qrtz9m6izUdnlLKHeZYI6V0T/ux3KrDXFe18+w6is/t+uMzyB4Ht+tbc4Sf9+qU0ndSSrenlJallG5NKf1jSukvxqpW6k8velbOUfWilNJfp5SuSCndn1J6KqV0Z0rpKymltTqsp6m9aOOVvudW3M4aq1qpP73oWTlH24tmD/Hcyiml/Tqop5G9qP2501NKl6SUHkgpPZhS+kFKafuxqJP+oBc9K+eof0dLKb0+pfTTlNKT7cfwiymlyR3W08heNMj33MDbp8aq3rE2oYc9bY9ExGvbt/8TEVtHxGUppTXGsYZvRcTuI1i/akT8Y7RqrZPPRERHDSQiIqX0oojYqf3haH4wquvjExHxroh4WUR8OSLeFhH/FhFHRMSZvSyKWtCLOrNuRPwiIj4crX/DVyLioIj4z9EmbHgvWuH/xJ+/714bEZ/tbTnUgF7Umc/Hs59Tr42I0yPimYj4wWgSNrkXpZReEq3HZXJEHBgRB7TfvySltFEva6Pn9KIOpJReGhGXRsR9EbFXRHwpIv4hIv6lg5yN7UXR+lqv3Lu/3I5d2KuiOtXxL+YN8EzO+Zr2+9eklH4TET+K1i/i/7Xy4pTSpIiYlHP+Q7cKyDnfExH3dCtfL6SUNo2Iv49WM/5Gh+ne3X57eUTsnVL6UDcf7x77cs55yYCPr0wpPRkR30wpbZRzvqtXhdFzelEHcs7zImLegLuuTCn9ISJmpZTWzTk/OIq0Te5FK9w64PsOIvSijuSc74iIOwbel1L6dkT8YKX//0eiyb3obyLi+RHxjpzzwxERKaWfRMTSaH3PdfozJf1LL+rMJyNicUTsk3N+JiIipZQi4isppS/nnBePImdje9FgX+uU0uci4pac83W9qapzjux5rgXttxtH/Olw3PkppRkppRsj4smI2L4d+8uU0lntQ06XpZQuTiltPjBZSuklKaXvp5SeSCktSikdsvKGgx0imFJaL6X0zZTS4vahd7emlD7aDj/WfvudAYeXraj3L1JKx6eU7k6tUxmuTym9baXcq6WUvp5Serhd+4kRsUoHj1lExL9GayJ6S4d5IlqT4muiNU1dOyLesvKC0Tw+KaVd0iCnmKWUrkwpnT3g49emlM5vH+74eErpupTS/l34d8UQP+j9ov32tRQasgAAIABJREFUhd3Yg8bQizr3QPvtqqP8/Mb2IhgBvagDKaVXRcSWEfHdDtI0uRetEq2jnn4/4L7ft+9LXdqDZtCLRmbriLhyxaCn7ZJoHeyx2yhzNrkXrfzvWDci3hyd9e6ec2TPc23cfvu7le47PiL+OVqHwv26/Q1wdbR+mfhgRCyL1gT1BymlzXLOT6SUUkScFxEviIj3R6sJ/VO0Tjf41VAFpJRWj4gro/XL/z9Fa4CyafsWEfHGaE1UvxARF7TvWzGdPTsiXhOtQ+TuiIh3RsT5KaXpA6aSx0XEIdE67eqmiPhAROw7SB2LotUkDh6q1va6t0XEDtE69PZ/V60tSSm9PCK2i9ZhhpdFxP3RaiznD1gz2sdn42GWsVFE/DgiTonW1+x10WpKy3POQz7hU/sc35zzLsPcZ4UdI2J5RNw6ws+j2TZuv9WLhtmL2msnRev/tr+K1ilJ5+Scf1f9WYPmmSi96Dvt76H7o/UDzWdyzk8Msz4mho3bb/WiEfSiAd7d/neeO4LPGbhn03vR3Gh9H30lpfTF9n1HR8RDMcjRG0xoG7ff6kXD60V/ERErH3XzVPvtX1V83qAmQC9a2T7RGrT197UMc84T9hYRx0TrMNHJ7dtmEXFFRDwaERu018yOiBwRW6/0uZ+PVhNZd8B960Tr/NIPtz9+W/tztx+wZqNo/bXiypXrGPDxodH65X/rIepes5334JXu37V9/1+vdP8PI+K/2u+vFxFPRMRRA+LPi9aTMa/0ebdHxLcLj+Gq0WqKH2p/vEu7hq1G+TU5OiL+OODx/7eIeDwi1ujC4zNobdFqSmcPkSu1vze+GRGXD7j/4HauNQfcd1lEXDbCf+/60WqWs3v9fHDr3U0v+lN81L1owNpb2nvniLgoIqaM8mvS6F4UERtExNcjYo92Pce0vx7n9fr54Na7m170p3jHvWjA59wREXM7+Jo0uhe1120drdMnVvTueyPif/f6+eDWu5te9Kd4J7+jzY2IBSvd9652HbNG8TVpfC9aKf/lKz9+/XhzGlfrifV0+3ZrRGwSEe/Kzz6P8bf5uefqvSlaF716NKU0ObWubP5YtA4xnN5e85qIuC/n/D8rPim3rsmyIKq9MSJ+McieJW+K1rT7xytqatd12YCaXhmtSe95A2paPvDjAfdvmnN+f2HPI6I1Wf3mCGsdyrsj4qoBj/93I2JKtH4hWWG0j8+wpJTWSSl9LaV0V/z5e2NmtP6jGVLOedec87CvTJ9SWjUivhetw5U/1kHJNINe1FkvWmHvaP2l54PtPf6r/Re8kWp0L8o5L845H55zPj/nfGXO+Zho9fM90hi+SgZ9QS/qTi+K1HpFqU2is9MAGt2LUkobROuIhwUR8db2bUFEXJBS+stu1E/f0os660XfiIhtU0qfSym9IKW0Q7SOHPpj+zZSje5FK+2zQUT8dfT5KVwRTuOKaE153xStCeDvIuLe3B7nDXDfIJ/3gmiduvSuQWKXtd+uOGpjZfdH62J0Q1kv/nzI30i8oL3n04PEVjyp1x9Qw8o1jUhKaWq0DjM8OCKe3/59asXL3D0/pbRGzvnxEeTbOlqHFX4rpbR2++4bo/UXnv3iz0+40T4+wzU7Wl/bz0frEMpHI+KwiNizWxu0f/k8PSJeERGvyzk/1K3c9C29aPCPRyTnfGP73Z+klG6OiKsi4g3R+gvNsEykXrSSsyPi3yNi24jo24sR0jG9aPCPR+Pd0fol84LSwsFMkF50ZLR+H9kn5/x0RERK6fJoHTX+f6L14h9MTHrR4B8PS875Bymlz0bE56J1mtvT7bd/H4M/bkOaIL1ooHdG68ih/9vlvOPOsKd1pff5hTUrN5aIiAejdY7i5weJrbj41O9i8IvuvjBah+kN5YH487mNI/FgRPw2ImZUrFlxnusL2+sH1jRSL4rWcOfsQWI/iVZDfdMI8q14Cb+vtG8DTU0prdMeioz28Xmy/Xbli7WuG61DRSOl9BfRemWIw3POp6xYkFLq9lFwJ0arMb0559yNi1rT//SiP3/cLT9vv90kRjDsiYnViwbKK71lYtKL/vzxqLWfq/tGxLl59NfBmgi9aIuIuHHFoCciIuf8h9S64O7LurQH/Ukv+vPHo5Jz/mJK6aSIeGm0TpWcFK3HZaSvwjkRetFA746Iq3POd49B7nHlNK7RuyxaR2XcmHOev9JtxYV2fxYR09qH8UZE6+rw0fqraSn3Nqn1Cg6DWXGxrb8Y5PPWj4jfD1LTimZ5Q7SeUH+agLafJKOZiN4erb+YD7ytOB3pfRHx8eEmah/p8q5onY+7cs73ROsCWXsP+HeO5vFZ8XJ6f7ooWUrpJREx8Or8q0WrET41YM3z49mHKHYkpfSpiPhIRLw353x1t/IyYelFQ3td++2vh/sJE6kXDWKf9tvSYewwGL3o2XaO1h/FRnUawATqRXdFxFbtU9tX5F8tIraKiEVd2oOJRS8aIOf8+5zzDe1hzIej9Zz7wXA/fwL1ohU5N47W0UN9fwpXhCN7OvHViHhvRFyeUjo5WtPaadE6v+/q3Loi+Pcj4vpoXTPiqGg9gf85yofjnR6tJ+MlKaVjonWe6ksjYrOc8yfbf/H4dUS8M6W0sJ33l9E6P/XiiLg0pfTlaB1et1a0Lnz3FznnT+WcH0gpzYqIf0opPdNe84H48+lXf5JSuj1a52YOek5ozvn30bpw1sDPWfHuz3LOCwfcPzsidsk5bzzEv3nHaF0Y7aic85UrB9sDkv2i9fLuo3p8cs73pJR+FhGfTykti9aw89MxYHqec36kvebolNKj0brI2CejdSjpWkPUvqLGy9o5hjwnNKX0nog4NlqHIf62ff7sCnfkwV+aHapM+F7UXnNGRNwWrdOPlkXrB7ZPRMRPo/UDyop1s0Mvina9z4/Wq1o8Gq1fTI+M1quX/bIqPwxBL3q2d0frr9yXDhbUi/7kW9F69aF5KaV/j9apEx+O1kXkZ1XlhyHoRa01m0ZrGHNttH7n/9to/TH+b/KAl2PXi57j3dG6UPdgZ670n1yDq0T36hYrXWF9iDWzI2L+ELENI+I70Trv8alo/QXiPyLiFQPW/GW0XhHmiWhNUg+N1jfPlVV1ROucx1Oj1XSejNaV2P9+QHy3aDWPJ6N1COPG7ftXi9ZL3d0erenp79r7/82Az10tWtdleCRaL215crQuzJlXqmFRjPBVomLoq6l/LyKurfi8r7frWX2I+Cfi2VeAH+3js2m0BlSPR6sB7RkrXem9veby9prftPd+1tcoBr/S+5UDv64V3095iNvBvX5OuPXmNlgPGGTN7NCLqh6fj0TriJRHonXR8xuidZ76miut04taa94dEfPb/9Y/tL9O/xwRq/X6+eDWu9tgPWCQNbNDLyo9jpMjYklEnFKxRi/687pdo/WqRA+2b1dF65fPnj8n3Hpz04u68nPRX7afV4+0n7tXRsROg6zTi57977kuIi7q9XOgW7fU/kfBmEqtq6YfnXOe0+tagIlLLwLqQC8C6kAvajbDHsZcSmnDaP0FeeOc8x9K6wHGgl4E1IFeBNSBXtR8hj0AAAAADeLVuAAAAAAaxLAHAAAAoEEMewAAAAAaZPJ4bpZScoEg6CM559TrGsaCXgT9RS8CamJpznlqr4voNr0I+s6welFHR/aklN6SUro1pXR7SumTneQCGC29CKgL/Qga7a5eFzBcehE02rB60aiHPSmlSRHxbxHx1ojYMiL2SyltOdp8AKOhFwF1oR8BdaAXARGdHdnzmoi4Ped8Z875DxFxVkTs2Z2yAIZNLwLqQj8C6kAvAjoa9rwoIu4e8PE97fueJaU0M6U0P6U0v4O9AIaiFwF1UexHehEwDvQioKMLNA92scTnXNwr5zwrImZFuPgXMCb0IqAuiv1ILwLGgV4EdHRkzz0R8ZIBH784Iu7trByAEdOLgLrQj4A60IuAjoY9P4uIl6eUXppSWjUi3h0R53enLIBh04uAutCPgDrQi4DRn8aVc34mpXR4RFwcEZMi4rSc841dqwxgGPQioC70I6AO9CIgIiLlPH6naDofFPpLznmwc777nl4E/UUvAmpiQc55eq+L6Da9CPrOsHpRJ6dxAQAAAFAzhj0AAAAADWLYAwAAANAghj0AAAAADWLYAwAAANAghj0AAAAADWLYAwAAANAghj0AAAAADWLYAwAAANAghj0AAAAADWLYAwAAANAghj0AAAAADWLYAwAAANAghj0AAAAADWLYAwAAANAghj0AAAAADWLYAwAAANAghj0AAAAADWLYAwAAANAghj0AAAAADWLYAwAAANAghj0AAAAADWLYAwAAANAgk3tdAACMl8mTq//bO+KII8aljjPOOKMyvnjx4mKOKVOmVMYPP/zwYo7jjz++uAYAgP7jyB4AAACABjHsAQAAAGgQwx4AAACABjHsAQAAAGgQwx4AAACABjHsAQAAAGgQwx4AAACABkk55/HbLKXx2wzoWM459bqGsaAX9af11luvMn7ggQcWc3z2s5+tjK+77rojqmm0Hnnkkcp4qc6IiH/+53+ujN9xxx3FHK9+9auLa+pALwJqYkHOeXqvi+g2vQj6zrB6kSN7AAAAABpkciefnFJaFBGPRcQfI+KZJk66gf6gHwF1oBcBdaAXAR0Ne9rekHNe2oU8AJ3Sj4A60IuAOtCLYAJzGhcAAABAg3Q67MkRcUlKaUFKaeZgC1JKM1NK81NK8zvcC6BKZT/Si4BxohcBdaAXwQTX6Wlcr8s535tSemFEXJpSuiXn/MOBC3LOsyJiVoQrvQNjqrIf6UXAONGLgDrQi2CC6+jInpzzve2390fEvIh4TTeKAhgp/QioA70IqAO9CBj1sCeltEZK6fkr3o+I3SJiYbcKAxgu/QioA70IqAO9CIjo7DSuaRExL6W0Is9/5pwv6kpVACOjHzXAIYccUhn/zGc+UxnfeOONu1jN6F1++eXFNcuWLauMX3fddcUcO+64Y2V8yZIlxRx0nV40hnbeeefimgMPPLAy/r3vfa+Y45JLLhl2TYPZYIMNims+//nPV8Zf9apXFXO86EUvqowfe+yxxRyXXnppZfy2224r5qCW9KKaW3PNNYtrrrzyysr4tGnTKuNnnnnmSEoaM2uttVZxzWGHHVYZz7l8pmHp/4irr766mKNpRj3syTnfGRH/u4u1AIyKfgTUgV4E1IFeBER46XUAAACARjHsAQAAAGgQwx4AAACABjHsAQAAAGgQwx4AAACABjHsAQAAAGiQNJzXrO/aZimN32ZAx3LOqdc1jAW9qPs22GCDyvisWbOKOXbZZZfK+JprrjmSkgZ11113VcbPOOOMYo6LL764Mj5//vxijieffLK4hj/Ti/rfOuusU1xz6qmnVsb33HPPYo6Uqr9VlixZUszx4x//uDK++eabV8bXW2+94h6lnjlebrjhhsr4zJkzizmuueaabpXTDxbknKf3uohum0i9qJ/Mnj27Mn7AAQeMTyF9Yu7cuZXxd77zneNUybgYVi9yZA8AAABAgxj2AAAAADSIYQ8AAABAgxj2AAAAADSIYQ8AAABAgxj2AAAAADSIYQ8AAABAgxj2AAAAADTI5F4XAEC9rbvuusU1F1xwQWV8m222KeZ44oknKuN33nlnZfzMM88s7nH66adXxm+//fZiDmDkDjzwwOKavffeu+N9PvShD1XGjzvuuDGvY/78+cU1s2bNqoyvttpqxRzbb799ZXzHHXcs5njlK19ZGf+P//iPYo5NN920uAYYubPOOqsy/va3v70yvvbaaxf3uOGGGyrjTz75ZDFHyUYbbVRc88IXvrDjfRYvXtxxjqZxZA8AAABAgxj2AAAAADSIYQ8AAABAgxj2AAAAADSIYQ8AAABAgxj2AAAAADSIYQ8AAABAg6Sc8/htltL4bdYAm222WXHNq171qsr4//t//6+YI6U07JoG89RTTxXXjOf3Gd2Tc+7sm6Om9KJnW3311SvjP/7xj4s5ttlmm8r4M888U8zxoQ99qDJ+6qmnFnPQTHpR/W200UaV8Z///OfFHOuuu27Hdfzud7+rjE+bNq2YY8GCBZXx/fbbrzJ+9913F/cYzs9Onfqv//qv4pp99tmnMv7ggw8Wc2y77baV8bvuuquYo48syDlP73UR3dakXjSRbLXVVpXxqVOnFnNce+21lfHHH3+8mOMVr3hFZfyiiy4q5thwww2La0q23377yvj8+fM73qNGhtWLHNkDAAAA0CCGPQAAAAANYtgDAAAA0CCGPQAAAAANYtgDAAAA0CCGPQAAAAANYtgDAAAA0CCGPQAAAAANknLO47dZSuO3WR9473vfWxk/8cQTiznWW2+9yviPfvSjYo6ddtqpuKbKQQcdVFxz9tlnV8afeOKJjmpgbOScU69rGAt60bOttdZalfFHHnmk4z3+8z//s7hm//3373gfmkkvqr+pU6dWxm+44YZijmnTpnWrnCH99Kc/La55+9vfXhl/4IEHulXOmNpwww2La3772992vM8b3/jGyvgVV1zR8R41siDnPL3XRXRbk3oR4+/uu++ujA+nF5XMmTOnuObwww+vjC9btqzjOmpkWL2oeGRPSum0lNL9KaWFA+5bN6V0aUrpV+2363RaLUCJfgTUgV4E1IFeBFQZzmlcsyPiLSvd98mIuCzn/PKIuKz9McBYmx36EdB7s0MvAnpvduhFwBCKw56c8w8j4sGV7t4zIlYcSzUnImZ0uS6A59CPgDrQi4A60IuAKpNH+XnTcs6LIyJyzotTSi8camFKaWZEzBzlPgAlw+pHehEwxvQioA70IiAiRj/sGbac86yImBXh4l9A7+hFQB3oRUAd6EXQfKN96fX7UkobRES0397fvZIARkQ/AupALwLqQC8CImL0w57zI2LF620fFBHndaccgBHTj4A60IuAOtCLgIiISDlXH7WXUvpuROwSES+IiPsi4h8j4tyI+F5E/GVE/CYi9s05r3xxsMFyTZhDBDfeeOPimgsvvLAyvvnmm3epmmoLFy6sjL/sZS+rjE+aNKm4xyabbFIZv/fee4s5mmSttdaqjJ9wwgnFHMccc0xlfPHixSMpaVA559Rxki7qVj+aSL1oOErfj4888kjHe5T+r4mIeOaZZzrepw5uuOGG4prTTz+9Mn7eeeWfzRctWjTckvqeXtT/vve97xXX7LvvvmNex6mnnlpcM3NmMy5j8va3v7245vzzz+94nze+8Y2V8SuuuKLjPWpkQc55eq+LWEEvolNbbLFFZfycc84p5ujG76w33XRTZfzVr351MceTTz7ZcR19ZFi9qHjNnpzzfkOEdh1xSQAd0I+AOtCLgDrQi4Aqoz2NCwAAAIAaMuwBAAAAaBDDHgAAAIAGMewBAAAAaBDDHgAAAIAGMewBAAAAaBDDHgAAAIAGmdzrAppq0aJFxTV/93d/VxnfYIMNijluvPHGyvh9991XzLHttttWxi+99NLK+NNPP13c449//GNxzUTy6KOPVsZzzsUchx56aGX8mGOOGUlJTGCl5/DChQuLObbaaqvKeEqpmGOVVVYprukHpZ46nDVHH310McesWbMq45/61KeKOWC8HHvsscU1++6775jXceutt475HnWxzTbb9LoEoIc222yz4povfOELlfHNN9+84zoeeuih4pqDDjqoMv7kk092XMdE5MgeAAAAgAYx7AEAAABoEMMeAAAAgAYx7AEAAABoEMMeAAAAgAYx7AEAAABoEMMeAAAAgAaZ3OsCJrJrrrlmzPd43vPK87yPfOQjHeV45JFHinvcd999xTUTyeTJ1U+9HXfcsZjjiSeeqIwfc8wxIymJLnvDG95QXHPttddWxh9//PFulVOp9L202267FXNMmzatMv7Zz362mGOttdYqrqmDCy64oDL+N3/zN8Uc22+/fWV83XXXLeY47LDDKuPf/OY3izkWLVpUXAPdcNtttxXXfO1rX6uMf/jDHy7mmDRpUmX8yCOPLOa47LLLKuPXXXddMQdAr5V+ToiI2GuvvTre56GHHqqM77rrrsUc119/fcd18FyO7AEAAABoEMMeAAAAgAYx7AEAAABoEMMeAAAAgAYx7AEAAABoEMMeAAAAgAYx7AEAAABokJRzHr/NUhq/zYiIiIMPPri45rTTTutoj3e84x3FNeeee25HezTN4YcfXhk/+eSTizm+9a1vVcY/8IEPjKimweScU8dJakgvotcOOeSQyvipp57a8R6f+MQnimtOOOGEjvcZD3oRERFHHnlkcc3xxx/f8T5z5sypjA/nZ6s6+NnPflZcM3369Mr4vffeW8yx5ZZbVsYfeeSRYo4+siDnXP2g9SG9qD/tsMMOlfHh/P41derUjuvYbrvtKuPXXXddx3vwHMPqRY7sAQAAAGgQwx4AAACABjHsAQAAAGgQwx4AAACABjHsAQAAAGgQwx4AAACABjHsAQAAAGgQwx4AAACABpnc6wL61fOeVz0n23///Ys53ve+91XGp0+fPqKaBjN5cudf4ttvv70yvvbaaxdz7LjjjpXxadOmFXM8/PDDlfFf/OIXxRyveMUrKuMvfelLizk23XTTyvhOO+1UzLHzzjtXxm+77bZijiOPPLK4ht558YtfXFyzePHiyvgf//jHbpVDzdx///29LgH6zimnnFJcs8cee1TGX//61xdzHHjggZXx2bNnV8avvPLK4h7d8E//9E+V8W78HHnWWWcV1zzyyCMd7wM81w477FAZnzdvXmV86tSpHdfw7//+78U1v/zlLzveh7FRPLInpXRaSun+lNLCAfcdk1L6bUrpuvbtbWNbJjDR6UVAXehHQB3oRUCV4ZzGNTsi3jLI/SfmnLdu377f3bIAnmN26EVAPcwO/QjovdmhFwFDKA57cs4/jIgHx6EWgCHpRUBd6EdAHehFQJVOLtB8eErpl+3DB9fpWkUAI6MXAXWhHwF1oBcBox72fCMiXhYRW0fE4oj4ylALU0ozU0rzU0rzR7kXwFD0IqAuhtWP9CJgjOlFQESMctiTc74v5/zHnPPyiDg1Il5TsXZWznl6zrnzlwQAGEAvAupiuP1ILwLGkl4ErDCqYU9KaYMBH+4VEQuHWgswVvQioC70I6AO9CJghcmlBSml70bELhHxgpTSPRHxjxGxS0pp64jIEbEoIg4dwxpr6SMf+Uhl/MQTTxynSsbepptuWhk/7bTTxqmS/vD4448X18yfX33E7Lvf/e5ijocffnjYNTVBv/Wivffeu7hm1qxZlfEnnniiW+UwjiZNmlRc87nPfW7M63jmmWfGfI+Jqt/6UVM89thjxTV77LFHZfyuu+4q5lhzzTUr45/61Kcq4z/60Y+Ke+ScK+PHHntsMcehh3b+Lfb1r3+9Mn7UUUd1vAdjRy/qX9tvv31xzbnnnlsZnzp1asd1nHzyyZXxT3ziE8Ucy5cv77gOxkZx2JNz3m+Qu789BrUADEkvAupCPwLqQC8CqnTyalwAAAAA1IxhDwAAAECDGPYAAAAANIhhDwAAAECDGPYAAAAANIhhDwAAAECDGPYAAAAANMjkXhfQr1772teO+R5PPvlkcc3VV19dGb/kkkuKOf7whz8Mu6bBbLjhhsU12267bUd7jJe77767uOb666+vjJ9//vnFHIsWLRpuSfSpk046qbjmTW96U2X8Bz/4QbfKoYu22267yviRRx5ZzDF9+vSO61i6dGll/MQTT+x4D+g3Dz30UGX8jW98YzHH7rvvXhnff//9K+M33nhjcY9f//rXlfG3vOUtxRy33HJLZfyII44o5vjOd75TXAN037ve9a7imqlTp3a0x8knn1xcc9RRR1XGO/09kd5yZA8AAABAgxj2AAAAADSIYQ8AAABAgxj2AAAAADSIYQ8AAABAgxj2AAAAADSIYQ8AAABAg6Sc8/htltL4bTbG1lhjjcr4NttsU8wxbdq0yvh///d/F3M89dRTxTUwWjnn1OsaxkJdetHixYsr49/4xjeKOU499dSO9uDZtttuu+Kaj3/845Xx/fbbr+M6Hn744eKavfbaqzJ+5ZVXdlxHXehFjKfjjjuuMn7UUUeNeQ3z588vrjnkkEMq49dff323yuHPFuScp/e6iG7Ti7rvda97XWX8oosuKuaYMmVKZfzee++tjO+0007FPRYtWlRcQy0Nqxc5sgcAAACgQQx7AAAAABrEsAcAAACgQQx7AAAAABrEsAcAAACgQQx7AAAAABrEsAcAAACgQQx7AAAAABpkcq8L6FePP/54Zfzqq68ep0qAfvWVr3ylMn7CCScUc3zwgx+sjJ966qnFHHPmzKmM33nnncUcdbD99tsX12y88caV8VNOOaWYY+211x5uSUNaunRpZfzMM88s5rjyyis7rgMmml133bW45hWveMU4VFJt1qxZxTXXX3/9OFQCrGyVVVYprvn+979fGZ8yZUoxx5IlSyrjb37zmyvjixYtKu5BszmyBwAAAKBBDHsAAAAAGsSwBwAAAKBBDHsAAAAAGsSwBwAAAKBBDHsAAAAAGsSwBwAAAKBBJve6AICJas6cOZXxpUuXFnOcdNJJlfGjjz66mOPAAw+sjJ9xxhnFHJdddllxTac23HDDyvjXv/71Yo511123W+UM6dprry2uKX1dLr744m6VA40xadKk4pqDDz64Mv5v//ZvxRzLly+vjJ9yyimV8fe85z3FPdZaa63K+Oabb17MAYyN6dOnV8b/5V/+pZhjzTXXrIyX+kxExKmnnloZv+WWW4o5mNiKR/aklF6SUroipXRzSunGlNI/tO9fN6V0aUrpV+2364x9ucBE9f/bu/d4q+py3+PfJ1BMyRJNJFIxjaNlO1SOutWMvBSoR+WIhpap2QbdkuLWY15KMbdtc4tmOzXxBroxFS/BzraXMDJsawKZYmhRQl4QRLwhXiAOqnOCAAAgAElEQVSe88cYKyfLtX6/udYcc43L/Lxfr/laa87fmL/xMNYaX+Z65hhjkkUAioAsAlAEZBGAmHpO41oj6TR330HS7pJOMrNPSTpT0kx3/6Skmel9AGgWsghAEZBFAIqALAIQFG32uPsSd5+Xfv+GpAWSBko6RFLbOQhTJB3arCIBgCwCUARkEYAiIIsAxHTpAs1mNkjSTpIekdTf3ZdISdhI2jzr4gCgI2QRgCIgiwAUAVkEoCN1X6DZzPpKukPSeHd/3czqfd4YSWO6Vx4ArIssAlAEZBGAIiCLAHSmriN7zGw9JSEy1d3vTB9eamYD0vEBkpZ19Fx3n+TuQ909fFlzAIggiwAUAVkEoAjIIgAh9Xwal0m6TtICd7+0ZmiGpGPS74+RND378gAgQRYBKAKyCEARkEUAYuo5jWtPSUdLesLMHksfO1vSRZJuM7PjJf1V0uHNKREAJJFFAIqBLAJQBGQRgCBz955bmVnPrQxAw9y9vhO/S6ZKWTR69Ojg+L/+679G59h2222zKqfwYv/nPfXUU9E5/ud//ic4/s1vfjM6x6pVq6LL4D1kESTpa1/7WnSZKVOmBMdXrlwZneOwww4Ljt93333B8RNPPDG6jiuvvDI4fscdd0TnGDVqVHQZZG5uFU97IovWNX78+OD4xIkTG17HNddcE13mhBNOaHg9qKy6sqhLn8YFAAAAAACAYqPZAwAAAAAAUCE0ewAAAAAAACqEZg8AAAAAAECF0OwBAAAAAACoEJo9AAAAAAAAFUKzBwAAAAAAoEJ6510AAKD7brnlluD43XffHZ3jsMMOC44PHz48OseBBx4YHO/bt290jtWrVwfHX3vtteD41KlTo+uYN29ecPzGG2+MzgGgObbffvvg+BVXXNHwOo499tjoMvfdd1/D62nUfvvtl3cJQCXFXvNI0vnnn9/0Op566qmmrwPgyB4AAAAAAIAKodkDAAAAAABQITR7AAAAAAAAKoRmDwAAAAAAQIXQ7AEAAAAAAKgQmj0AAAAAAAAVQrMHAAAAAACgQmj2AAAAAAAAVEjvvAsAADTPG2+8EV1m8uTJDY1L0i677BIc32effaJzLFq0KDg+bdq06BwAyuull14Kjr/77rsNr+Okk06KLnPHHXcEx/v16xcc33XXXbtUU0fuvvvuhucAWlFs/zzvvPOic/Tt2zercoBccWQPAAAAAABAhdDsAQAAAAAAqBCaPQAAAAAAABVCswcAAAAAAKBCaPYAAAAAAABUCM0eAAAAAACACqHZAwAAAAAAUCG98y4AAFB+c+fObWgcAF5++eXg+Pbbbx+d48YbbwyO77///tE5fv/73wfH165dGxwfMmRIdB0xd955Z8NzAK3o+9//fnD805/+dI/UMXbs2OD47bff3iN1oLVxZA8AAAAAAECF0OwBAAAAAACoEJo9AAAAAAAAFUKzBwAAAAAAoEJo9gAAAAAAAFQIzR4AAAAAAIAKodkDAAAAAABQITR7AAAAAAAAKqR33gUAAAAAMS+99FJ0mREjRgTH99tvv+gcX/7yl4Pjxx9/fHDc3aPruO6664LjDz30UHQOAO930003Bce//vWvR+f4wx/+EBy/4IILonNMmzYtOF5PTgCNih7ZY2ZbmtkvzWyBmT1pZqekj08ws+fN7LH0dkDzywXQqsgiAEVAFgEoArIIQEw9R/askXSau88zsw9Jmmtm96djl7n7Jc0rDwD+jiwCUARkEYAiIIsABEWbPe6+RNKS9Ps3zGyBpIHNLgwAapFFAIqALAJQBGQRgJguXaDZzAZJ2knSI+lD48zscTO73sw26eQ5Y8xsjpnNaahSAEiRRQCKgCwCUARkEYCO1N3sMbO+ku6QNN7dX5d0laRtJQ1R0lWe2NHz3H2Suw9196EZ1AugxZFFAIqALAJQBGQRgM7U1ewxs/WUhMhUd79Tktx9qbv/zd3XSrpG0q7NKxMAyCIAxUAWASgCsghASD2fxmWSrpO0wN0vrXl8QM1iIyXNz748AEiQRQCKgCwCUARkEYAYc/fwAmZ7Sfq1pCckrU0fPlvSkUoOD3RJiySNTS8UFporvDIAheLulncNbcgioHWRRQAKYm5RTnsii4CWVlcWRZs9WSJIgHIp0h9YWSKLgHIhiwAURGGaPVkii4DSqSuLuvRpXAAAAAAAACg2mj0AAAAAAAAVQrMHAAAAAACgQmj2AAAAAAAAVAjNHgAAAAAAgAqh2QMAAAAAAFAhNHsAAAAAAAAqhGYPAAAAAABAhdDsAQAAAAAAqBCaPQAAAAAAABVCswcAAAAAAKBCaPYAAAAAAABUCM0eAAAAAACACqHZAwAAAAAAUCE0ewAAAAAAACqkdw+vb7mkxTX3N0sfK7qy1CmVp1bqzFYz6tw64/mKpKxZJJWnVurMVivXSRYVU1lqpc5stXqdVc2j9lkk8bPOGnVmqyx1Sjm+NjJ3z3i99TOzOe4+NLcC6lSWOqXy1Eqd2SpLnUVVpu1XllqpM1vU2RrKtP3KUit1Zos6W0dZtiF1Zos6s5dnrZzGBQAAAAAAUCE0ewAAAAAAACok72bPpJzXX6+y1CmVp1bqzFZZ6iyqMm2/stRKndmiztZQpu1XllqpM1vU2TrKsg2pM1vUmb3cas31mj0AAAAAAADIVt5H9gAAAAAAACBDNHsAAAAAAAAqJLdmj5kNN7OnzWyhmZ2ZVx0xZrbIzJ4ws8fMbE7e9bQxs+vNbJmZza95rJ+Z3W9mf0q/bpJnjW06qXWCmT2fbtfHzOyAnGvc0sx+aWYLzOxJMzslfbxw2zRQa6G2aVmQRY0rSx6VIYvSmkqRR2RRtsiixpFF2SKLWhNZ1DiyKFtkUQM15XHNHjPrJemPkvaX9JykRyUd6e5/6PFiIsxskaSh7r4871pqmdneklZKutHdd0wfu1jSCne/KA3nTdz9W3nWmdbVUa0TJK1090vyrK2NmQ2QNMDd55nZhyTNlXSopGNVsG0aqPUIFWiblgFZlI2y5FEZskgqTx6RRdkhi7JBFmWLLGo9ZFE2yKJskUXdl9eRPbtKWujuf3H3dyXdIumQnGopJXd/UNKKdg8fImlK+v0UJb9cueuk1kJx9yXuPi/9/g1JCyQNVAG3aaBWdB1ZlIGy5FEZskgqTx6RRZkiizJAFmWLLGpJZFEGyKJskUXdl1ezZ6CkZ2vuP6fihrJLus/M5prZmLyLiejv7kuk5JdN0uY51xMzzsweTw8hzP1QxjZmNkjSTpIeUcG3abtapYJu0wIji5qn0PtOO4Xdb8qSR2RRw8ii5insftOBwu43ZFHLIIuap7D7TQcKu9+QRV2TV7PHOnisqJ8Bv6e77yxphKST0sPd0LirJG0raYikJZIm5ltOwsz6SrpD0nh3fz3vekI6qLWQ27TgyCIUdr8pSx6RRZkgi1DY/YYsailkEQq735BFXZdXs+c5SVvW3P+4pBdyqiXI3V9Ivy6TdJeSwxuLaml6rmDbOYPLcq6nU+6+1N3/5u5rJV2jAmxXM1tPyY451d3vTB8u5DbtqNYibtMSIIuap5D7TntF3W/KkkdkUWbIouYp3H7TkaLuN2RRyyGLmqdw+01HirrfkEXdk1ez51FJnzSzbcxsfUmjJc3IqZZOmdlG6cWVZGYbSfqipPnhZ+VqhqRj0u+PkTQ9x1qC2nbM1EjlvF3NzCRdJ2mBu19aM1S4bdpZrUXbpiVBFjVP4fadjhRxvylLHpFFmSKLmqdQ+01nirjfkEUtiSxqnkLtN50p4n5DFjVQk+fwaVySZMlHjv1AUi9J17v7hbkUEmBmn1DSKZak3pJuLkqdZvYTScMkbSZpqaTzJP1U0m2StpL0V0mHu3vuF93qpNZhSg5lc0mLJI1tO+cyD2a2l6RfS3pC0tr04bOVnGdZqG0aqPVIFWiblgVZ1Liy5FEZskgqTx6RRdkiixpHFmWLLGpNZFHjyKJskUUN1JRXswcAAAAAAADZy+s0LgAAAAAAADQBzR4AAAAAAIAKodkDAAAAAABQITR7AAAAAAAAKoRmDwAAAAAAQIXQ7AEAAAAAAKgQmj0AAAAAAAAVQrMHAAAAAACgQmj2AAAAAAAAVAjNHgAAAAAAgAqh2QMAAAAAAFAhNHsAAAAAAAAqhGYPAAAAAABAhbR0s8fMJpiZ19xeMLM7zGzbJq/3djOb1a6O5V14/vrpc4ZkWNM4M/NuPO+jZvZDM/utmb1rZosyqudX6c9k3248N/Pt08E6jk3r69vF5/1vM7vBzBaa2Soze9rMzjOzDZpVK4qPLFpnzm5lUfrcvczsf8zs7XQbXmhmvRusp5JZlD53qJndZ2Yvm9kKM/uFme3WjDpRDmTROnN293XRfmZ2q5ktTv+fn5/O1avBeqakP5Pju/n8M8xsWCM1ROYflta3Yxefd6yt+ztXe7u6WfWi2MiidebsbhYN62S/uqjBeir7uih9/v81s0fN7K309dE9ZrZR1nX2lJZu9qRek/SP6e10SUMkzezhH+q1kr7UheXXl3SeklrzNlDSlyW9KOmxLCY0s4GSPpfePbIbUxRp+7T3ZUnbSvq+pAMkXSHpXyRNzbMoFAJZ1AAz20bS/ZKWShop6d8knSLpkgbmrGwWmdmWkn4hqbekr0k6Ov3+PjPbOs/akDuyqDFjJG0k6dtK/p+/RdJESRd3d8L0DaFD07vdySJJOkPSsO7W0ER3673ft7bb+HTsv/MqCoVAFmXjK1p3/7qiuxNV+XWRJJnZNyTdrCR7Rkj6hqQ/KXl9VEqlLTxDa9z94fT7h83sr5J+reQ/6GntF07fmenl7u9mVYC7Pyfpuazm62GPu3t/STKzSySNymDO0enXByQdZmb/nOX2ztn33f2lmvuzzOxtSVeb2dbuvjivwpA7sqgxZ0paImmUu6+RJDMzSRPN7PvuvqQbc1Y5iw6U9CFJ/9fdX5UkM/uNpOVKfueuyrE25Issasw/u3vtkQCzzGxDSaea2dnu/k435jxQ0saSZkr6gplt4e4vZlFs3tLXRLWvi2RmRyv5Q59mT2sji7LxuLvPz2iuyr4uMrPNJF0m6Zvufk3N0F05lZQJjux5v7np10GSZGaTzWyOmR1qZk9KelvSbunYVmZ2S3r4+yozu9fM/lftZGa2pZn9PD0UbFHaMVS7Zd53iKCZbWpmV5vZEktOSXjazNre6Xgj/XpDzSF5bfVuYGYXm9mzZvaOmf3ezA5oN3cfM/uRmb2a1n6ZpPW6s7HcfW13nhdxpKSHlRz98hFJw9sv0J3tY50cYmxms8zs9pr7/2hmM9JDRt80s8fM7CtZ/MPaNXra/C79unkW60BlkEVdM0TSrLZGT+o+JW9qfLGbc1Y2i5Rs5zWSVtY8tjJ9zDJaB6qBLOqCdo2eNr+TtIGShk13HCnpeUnjlLx2P6L9Amb2wfTfuTj9dz5jZv+Wji2StKmk82q2z7A0j9zMDmo312Qzm1Nzf/v05/ps+nN90szGm1nmf0dY8gf7KEl3drMxhuoii/JX5ddFbbk6JaP5CoEje95vUPr1xXaPXSzpu0pOEXjGzPpJmi3pZUknSFql5J3lX5jZYHd/y8xM0nRJm0k6XkkInS+pn5JDwjpkZh+UNEvJH//nS3pK0nbpTZL2UdJR/Vclh79KyTvaknS7pF2VHCL3ZyW/uDPMbKi7t51mdZGSw9LOkfQHSf8k6fAO6lik5I+nYzurNWtm9klJuyg5/WKmpGVKgmVGzTLd3T6D6ixja0kPSfqxkp/ZnkpCaa27/yRQ+yxJcvdhda6nzR6S1kp6uovPQ7UNSr+SRfVl0QaS2r+71PaHwg6B53WoBbLoDiW/RxPN7ML0sXMlvaIO3jFFSxuUfiWLuv+6aA9Jyzt5wyfIzD6k5Mieq9z9KTObpySLflizTNt2/UdJFyj5o7j2dIuRkn6pZFtcmz72ByXbvR4DlbxGmarkj7UhSn4OH1RyymxntU+WNMzdB9W5HknaV8nPudOMQ8salH4li7qWRQ+Y2aaSnlWy//+bu/+tjue1X2fVXxftpiTnjjezcyT1lzRP0qnu/ps66ysed2/Zm6QJSg5Z753eBiv5z/B1SQPSZSZLcklD2j33AiUh0q/msU2UHHZ6Unr/gPS5u9Uss7WSd05nta+j5v5YJX/8D+mk7r7pvMe2e3zf9PHPt3v8QUnT0u83lfSWpG/VjH9Ayc7o7Z63UNJ1Xdiel0ha1ODP5FxJf6vZ/ldIelPSRhlsn2Hp4zu2e3yWpNs7mcvS342rJT1Q8/ix6Vx9ax6bKWlmF/+9WygJy8l57w/c8ruRRX8f73YWKWlezG332JfTOiZ142dS+SxS8gfbc+nzXdILkj6b9/7ALb8bWfT38UxeF6XP+VQ6/4Ru/ky+lv4bdk3v/7/0/jY1y3wpfezgwDzL29eg5A8sl3RQu8cnS5rTyTxtWXS2pL/UPP6+XJN0naSFXfz3Xq/kj/Zeee8P3PK7kUV/H2/kddFOSpqxB0jaT9IPlLyuubybP5NKvy6SdK+SZvbzSq5zNFxJY+p1Sf3z3ie6e+M0rmTHWp3enpb0CUlf9nWv7/C8v9dxbbOfkouBvm5mvS35xJc3lLybMjRdZldJS939kbYneXJNlrkK20fS7zpYZ8x+SrrdD7XVlNY1s6amzyh5B3x6TU1ra+/XPL6du3frUx8aMFrSr2q2/08kbSjp4Jplurt96mJmm1jyCWOL9d7vxhgl/9F0yt33dfe6r0xvZutLuk3JqROnNlAyqoEsaiyLrpK0s5l9x8w2M7PdlbxD9rf01lWVziIzG6DkXca5Si5COCL9/m4z2yqL+lFaZFFGr4vMbBMljejHJX2vi7W3OVJJU+W36f1blPwhM7pmmX0krXD3Ge2fnIX09JPzzWyhkiMmV0u6UNI2FvjEQ3c/3t2362y8g/Wsr+QopGnejSMPUDlkUQNZ5O6/c/ez3P3n7v4Ldx+v5I35f7bk+jRdVenXRUoaa30lHe/uU939HiUXxv+bklNoS4nTuJIu735K/uN8UdILnrb3aizt4HmbSdpdyTvH7c1Mv7YdtdHeMiUXxuzMpnrvkL+u2Cxd5+oOxtr+09yipob2NeXKko/h20HStWb2kfThJ5W823yk3jukt7vbp16TlfxsL1ByCOXrkk6UdEhWK0gPH71R0qcl7enur2Q1N0qLLOr4fl3c/Rdm9m1J31FyOPfq9OvJ6ni7dapFsuj/KXkNMMrdV0uSmT2g5PD105VsN7Qmsqjj+11iySdoTZfUR8kRN12+iGn6B9l+kq6oyaI3JD0q6Si9dwpVs7Po+0pOLTlfyWkNryrJoW8r+eN0ZedP7ZIRSq4DwilckMgidXK/Ebcr+WS+f1By1EpdWuR10Yr066y2B9z9dTObq+QIzVKi2ZNc6X1OZJn2wSIlvxAzlPyytdd28akX1fFFdzdXcpheZ17We+c2dsUKJYeeHRpYpu0818313i912/28tX2E38T0VuujZrZJ2hTp7vZ5O/26frvH+yk5VLTtxdmBksa5+4/bFrDsL0J4mZJg2t/dn8p4bpQTWfTe/W5x9wvN7HJJ2yg5PamXku3ycPCJ79cKWbS9pCfbGj2S5O7vWnKRy20zWgfKiSx67363WHKR4ZuVvKGzh7t3qeFc43Alr9VPSW/t17OjJ5+y87KkAd2YP5RF7ev4D3f/+8fHm9mB3VhfzGhJf5VU3utjIEtk0Xv3s9bRdgtphddFC5Rsl/YfUmFKTk0rJU7j6r6ZSv4Tf9Ld57S7tV1o91FJ/c1st7YnpYfH71zH3DuZ2T90Mt727tAGHTxvC0krO6ipLSyfULJD/b0Dmu4kmR210h3pkS5fVnI+7hfa3Y5SciX6w9LFu7t92j468e8XazWzLSXVXp2/j5I/EN+pWeZDWvcQxYaY2VmSvinpq+4+O6t50bLIohruvtLdn0hfdJwkabGkX9T7/BbKosWSdkxPm2ibv4+kHSUtymgdaC1k0XuuVHK9h/9T82/vjiOV/AHSPouGKzlCoO1UrpmS+lm7T9Vq5129f/ssS+epzaK+Si70XOuDWjeLemnd08gaZsnH0/8fSbd0cPQG0BVkUecOU3JdosfrfUILvS76mZLGzhdq5v+wkotS/z6jdfQ4juzpvkslfVXJFc7/Q0m3tr+kz0ua7ckVwX+u5Jdjmpl9S8kO/F3FD8e7UckfKfeZ2QQl56luI2mwu5+Zvvv6jKQjzGx+Ou/jSs5PvVfS/Wb2fSWH122s5CKcG6Tnbb5sZpMknW9ma9Jl/knJOYrrSM/N/lXsnFAzG5V+O1jShjX3f+XpJ09Y/BMZ9lByYbRvufusDtZxlpIXPdd2d/u4+3Nm9qikC8xslZJm59mq6Z67+2vpMuea2etKOrlnKjmUNPiRqWY2M52j03NCzewoJeftT5b0fHpdkTZ/9m58UgdaHlmULLOdkhcdv1Xyf9tBkr4u6UCv+Th2sujvrlVyWsZdZnalkhc4Jyk5OmBSaH6gE2RRsszZSq4h8W+S1rb7f/4P7v56utwESee5e/t3kdvm+bikvSSd1UkW3aMki75d8++82cy+q+RUqwGS9nb3selTnpJ0YPq8lZKedvc3zGy6pFMtuQbGq5JO0/uPbLhf0knpv3+Fkp9Fn862QU2N1ym5IG097/QfLGkjcQoXGkcWJctcJeklJY2td5VcqHmcpB+4+8s1y00Wr4vk7nPSPLzOzM5UckTRGUoa4leE5i80L8BVovO6qd0V1jtZZrI6/0SCj0m6Qcn5ou8oeTf0PyV9umaZrSTdo+Q/zsVKrlJ+uwJXek8f21TSNUpC520l/0mfXDP+RSXh8baSQ84GpY/3UXJO9UIlO/aL6foPrHluHyXvOr2m5GN2/0PSv+j9V3pfpDo+JUrvfZJL+9uwmmVuk/TbwBw/Suv5YCfjZ2jdK8B3d/tsp+RczDeVBNAhanel93SZB9Jl/pque52fkTq+0vus2p9r4Peps+11bN77BLd8bh1lQAfLTBZZFNo+Wyn5VIvX0n13lqTPdbAcWfTecvum22xFevuVanKbW+vdOsqADpaZLLIotH1mqb7XRRdLWhaY5/Q0az7eyfgRqvk0ISVH31yi5B3ydyQ9I+nCmuV3UXJK65u1tSj5I3i6kmtfLFbSqFrnZ5wuc1e6zNK09n+qzR51/Glck1Xnp7RK+qmkp/LeB7gV40YWZZJFJ6d1vJFugycljZf0gXbL8broveX6KvnAj5fT34tfSPpM3vtDIzdL/2FAU6XvGJ3r7lPyrgVA6yKLABSBmf1KyccFn593LQBaF6+Lqo1mD5rOzD4maY6Szm2XP40CALJAFgEoAks+cnmppO2d07cB5ITXRdVHswcAAAAAAKBC+DQuAAAAAACACqHZAwAAAAAAUCE9+tHrZsY5Y0CJeCcfCVt2ZBFQLmQRgIJY7u4fzbuIrJFFQOnUlUUNHdljZsPN7GkzW5h+Hj0A9DiyCEBRkEdApS3Ou4B6kUVApdWVRd1u9phZL0lXSBoh6VOSjjSzT3V3PgDoDrIIQFGQRwCKgCwCIDV2ZM+ukha6+1/Sj2q7RdIh2ZQFAHUjiwAUBXkEoAjIIgANNXsGSnq25v5z6WMA0JPIIgBFQR4BKAKyCEBDF2ju6GKJ77u4l5mNkTSmgfUAQAhZBKAoonlEFgHoAWQRgIaaPc9J2rLm/sclvdB+IXefJGmSxJXeATQFWQSgKKJ5RBYB6AFkEYCGTuN6VNInzWwbM1tf0mhJM7IpCwDqRhYBKAryCEARkEUAun9kj7uvMbNxku6V1EvS9e7+ZGaVAUAdyCIARUEeASgCsgiAJJl7zx21xyGCQLm4e0fnfJceWQSUC1kEoCDmuvvQvIvIGlkElE5dWdTIaVwAAAAAAAAoGJo9AAAAAAAAFUKzBwAAAAAAoEJo9gAAAAAAAFQIzR4AAAAAAIAKodkDAAAAAABQITR7AAAAAAAAKoRmDwAAAAAAQIXQ7AEAAAAAAKgQmj0AAAAAAAAVQrMHAAAAAACgQmj2AAAAAAAAVAjNHgAAAAAAgAqh2QMAAAAAAFAhNHsAAAAAAAAqpHfeBQAAAAAAgGKZMGFCdJmDDjooOD506NCMqkFXcWQPAAAAAABAhdDsAQAAAAAAqBCaPQAAAAAAABVCswcAAAAAAKBCaPYAAAAAAABUCM0eAAAAAACACqHZAwAAAAAAUCE0ewAAAAAAACqkd94FtLKDDjooOP6Vr3wlOscRRxzRcB0vv/xycHzq1KnB8SlTpjS8jmeffTY6BwAAAAAgG3379g2Of/WrX43OsemmmwbHd9ttt+gcf/7zn4Pjy5cvj86B9+PIHgAAAAAAgAqh2QMAAAAAAFAhNHsAAAAAAAAqhGYPAAAAAABAhdDsAQAAAAAAqBCaPQAAAAAAABVCswcAAAAAAKBCzN17bmVmPbeyErjggguC42eddVYPVdJ8b731VnD81FNPjc4xe/bs4PhTTz3VpZoQ5+6Wdw3NQBYB5UIWtYahQ4cGxz/60Y82vI4HH3wwukxsPTvssEPDdYwdOzY4PmjQoOgcTzzxRMN1zJgxIzg+bdq0htdRMXPdPfyLWkJkUesaMWJEcPy4446LzjFq1KiG6xg5cmRwfPr06Q2vo2LqyqLejazBzBZJekPS3yStqWL4ASgH8ghAEZBFAIqALALQULMn9QV3X57BPADQKPIIQBGQRQCKgCwCWhjX7AEAAAAAAKiQRps9Luk+M5trZmOyKAgAuok8AlAEZBGAIiCLgBbX6Glce7r7C2a2uaT7zewpd1/nyndpuBAwAJotmEdkEYAeQhYBKAKyCGhxDR3Z4+4vpF+XSbpL0q4dLDPJ3YdyUTAAzRTLI7IIQE8giwAUAVkEoNvNHjPbyMw+1Pa9pC9Kmp9VYQBQL/IIQBGQRQCKgCwCIDV2Gld/SXeZWfau85MAABvoSURBVNs8N7v7PZlUBQBdQx4BKAKyCEARkEUAut/scfe/SPpshrWgwjbccMPg+I9//OPoHK+88kpw/D//8z+jc5x66qnRZVA+5FE1rLfeesHxb3zjG8HxK664IrqO9IVvp9asWROdo1evXsHxefPmRee47bbbguNXXnlldI6VK1dGl0HPIosac9RRRwXHR4wYEZ1j8ODBwfHXXnstOkcsi2KvaepZx0MPPRQcv/baa6NzfOlLXwqO77HHHtE5jjjiiOD4q6++Gp3j/vvvjy6DnkUWoV6bb755cHyXXXZpeB333BPvM06fPr3h9eD9+Oh1AAAAAACACqHZAwAAAAAAUCE0ewAAAAAAACqEZg8AAAAAAECF0OwBAAAAAACoEJo9AAAAAAAAFUKzBwAAAAAAoELM3XtuZWY9t7IS+PCHPxwcnz59enSOwYMHB8evuuqq6BwTJ06MLtOoz3/+88Hxiy++ODrHDjvsEByv53f5mWeeCY7vu+++0TmeffbZ6DJV4e6Wdw3NQBYV03nnndfQ+KOPPhpdxyWXXBIc/+Mf/xid4xvf+EZw/KSTTorOEcur559/PjrH6aefHhy/7bbbonOUBVkESdpoo42iywwYMKAHKglbtWpVdJkXXnih6XXcfffd0WX23nvv4PiIESOic8yePbvumipgrrsPzbuIrJFFrevxxx8Pju+4447ROd55553g+MiRI6Nz3HPPPdFlsI66sogjewAAAAAAACqEZg8AAAAAAECF0OwBAAAAAACoEJo9AAAAAAAAFUKzBwAAAAAAoEJo9gAAAAAAAFQIzR4AAAAAAIAKodkDAAAAAABQIb3zLqCVvfbaa8HxYcOG9UwhPeC///u/g+Njx46NzrHDDjsEx80sOscnPvGJ4PjXv/716Bznn39+dBkA6zrxxBOjy4wfPz44ftJJJwXHb7755ug6Yrlbjy222CI4vnLlyugcvXuH//vddttto3NMnDgxOP7II49E51i8eHF0GaAo3nzzzegyCxcu7IFKimGDDTYIjm+11VbROc4555zg+OzZs7tUE4DiGDFiRHSZwYMHN7ye2N9G99xzT8PrQPdwZA8AAAAAAECF0OwBAAAAAACoEJo9AAAAAAAAFUKzBwAAAAAAoEJo9gAAAAAAAFQIzR4AAAAAAIAKodkDAAAAAABQIb3zLgCt4aqrrgqO77XXXj1USdh2222XdwlAKe2yyy7B8csvvzw6xz777BMcnz17dpdqapZTTjklOH7FFVdE59h4442D46tXr47OcfLJJ0eXAVBd48aNC45/6lOfis5x++23Z1UOgIz16dMnOH7UUUcFx7/1rW9F17H++usHx2+99dboHJdcckl0GeSDI3sAAAAAAAAqhGYPAAAAAABAhdDsAQAAAAAAqBCaPQAAAAAAABVCswcAAAAAAKBCaPYAAAAAAABUCM0eAAAAAACACqHZAwAAAAAAUCG98y4Axde7d/jX5IQTTojOMWbMmOC4u3eppu5au3ZtcPzqq6/ukTqAqvngBz8YHP/Tn/4UnePhhx/Oqpym2mqrrYLjw4cPj85xwAEHZFUOgAraYostosucccYZwfFx48ZF53jhhRfqrglAz1qzZk1w/JhjjgmODx48OLqOBQsWBMfr+dsoVifyEz2yx8yuN7NlZja/5rF+Zna/mf0p/bpJc8sEAPIIQDGQRQCKgCwCEFLPaVyTJbV/m/JMSTPd/ZOSZqb3AaDZJos8ApC/ySKLAORvssgiAJ2INnvc/UFJK9o9fIikKen3UyQdmnFdAPA+5BGAIiCLABQBWQQgpLvX7Onv7kskyd2XmNnmnS1oZmMkhS/YAgDdV1cekUUAmowsAlAEZBEAST1wgWZ3nyRpkiSZWc9chRcA2iGLABQBWQSgCMgioPq6+9HrS81sgCSlX5dlVxIAdAl5BKAIyCIARUAWAZDU/WbPDEltn/V2jKTp2ZQDAF1GHgEoArIIQBGQRQAk1XEal5n9RNIwSZuZ2XOSzpN0kaTbzOx4SX+VdHgzi0S+jj766OD4D37wgx6qJGzNmjXRZc4555zg+OzZs7MqB01AHhXXwoULg+NLly6NzuFejqPIzzwz/MEmRx11VHSOWbNmZVQN8kAWodl+9KMfRZdZsaL9dXnXde2112ZVDgqKLKq2c889Nzi+9957B8dXr14dXUfs7zher5RbtNnj7kd2MrRvxrUAQBB5BKAIyCIARUAWAQjp7mlcAAAAAAAAKCCaPQAAAAAAABVCswcAAAAAAKBCaPYAAAAAAABUCM0eAAAAAACACqHZAwAAAAAAUCE0ewAAAAAAACqkd94FIH8bb7xxcPzkk0/uoUoa8+ijj0aXmThxYg9UArSeF198MTh+6623Rue48cYbg+Nf+cpXulRTRz784Q8HxydMmBCd47XXXguO33XXXV0pCUAL2muvvYLjBx10UHSOAw88MDi+evXqLtUEIDvbb799cPzSSy+NzrH77rs3VMNXv/rV6DLTpk1raB0oNo7sAQAAAAAAqBCaPQAAAAAAABVCswcAAAAAAKBCaPYAAAAAAABUCM0eAAAAAACACqHZAwAAAAAAUCE0ewAAAAAAACqkd94FIH+HHHJIcPwzn/lMD1US9uSTTwbHR44c2UOVAOiqq6++OrrMoYceGhw/8sgjg+O33nprdB0XX3xxcLxfv37ROc4777zoMgAQcu211wbHX3nllegcCxcuzKocABkbMmRIcHz48OENr+Pf//3fg+PTpk1reB0oN47sAQAAAAAAqBCaPQAAAAAAABVCswcAAAAAAKBCaPYAAAAAAABUCM0eAAAAAACACqHZAwAAAAAAUCE0ewAAAAAAACqkd94FoLk+8IF4P+/AAw8MjptZ0+tYtWpVdI7DDz88OL58+fIu1QSgWI499tjg+L333hscHzduXHQdL7/8cnD84IMPjs4BACGx1yuSNGjQoOD4mDFjonMsXry43pIAZGjvvfeOLjNkyJDg+E033RSd46GHHgqO33DDDdE5imCzzTaLLvO5z30uOP7SSy9F55gzZ05w/O23347OUTUc2QMAAAAAAFAhNHsAAAAAAAAqhGYPAAAAAABAhdDsAQAAAAAAqBCaPQAAAAAAABVCswcAAAAAAKBCaPYAAAAAAABUCM0eAAAAAACACumddwForg033DC6zKhRo4Lj7t5wHWvXrg2On3XWWdE5/vjHPzZcB4DiWrp0aXD88ssvD45fd9110XUcd9xxXaoJANobOHBgcPy2226LzhF7bfWb3/ymSzUB6DkHH3xwdJm33norOH7qqadG51ixYkXdNeVp7733Do7feuut0Tn69+/fcB2nnXZacPyyyy5reB1lEz2yx8yuN7NlZja/5rEJZva8mT2W3g5obpkAWh1ZBKAoyCMARUAWAQip5zSuyZKGd/D4Ze4+JL39PNuyAOB9JossAlAMk0UeAcjfZJFFADoRbfa4+4OSynEMGYDKIosAFAV5BKAIyCIAIY1coHmcmT2eHj64SWcLmdkYM5tjZnMaWBcAdIYsAlAU0TwiiwD0ALIIQLebPVdJ2lbSEElLJE3sbEF3n+TuQ919aDfXBQCdIYsAFEVdeUQWAWgysgiApG42e9x9qbv/zd3XSrpG0q7ZlgUAcWQRgKIgjwAUAVkEoE23mj1mNqDm7khJ8ztbFgCahSwCUBTkEYAiIIsAtOkdW8DMfiJpmKTNzOw5SedJGmZmQyS5pEWSxjaxRjTge9/7Xt4l1OWZZ57JuwQUHFlUff369QuOjxo1Kjj+yiuvRNfxsY99rEs1AR1ptTzabLPNosvcddddwfHtt9++4TrMLLqMuze8npg+ffo0XENsmd/+9rfROd56663g+O233x6dIwvTpk0Ljs+ePbtH6mhFrZZFRXH66afnXUKPGT16dHSZnXfeOTjev3//6ByxfF+9enV0jueeey66TKuJNnvc/cgOHr6uCbUAQKfIIgBFQR4BKAKyCEBII5/GBQAAAAAAgIKh2QMAAAAAAFAhNHsAAAAAAAAqhGYPAAAAAABAhdDsAQAAAAAAqBCaPQAAAAAAABVCswcAAAAAAKBCeuddAJrr8MMPz7sESdLNN98cHP/5z3/eQ5UAKKrvfe97wfG+ffsGxy+66KLoOrbbbrsu1QRAWr58eXSZQw45JDjer1+/rMppSCxHJOnyyy8Pju+1117B8SlTpkTXsWjRougyPWG//fYLju+8887ROTbYYIPg+OzZs7tUE1B0e+65Z3SZOXPmBMffeeedrMppyIknnhgcP/XUU6NzZPHa6plnngmOP/zww9E5pk2b1nAdVcORPQAAAAAAABVCswcAAAAAAKBCaPYAAAAAAABUCM0eAAAAAACACqHZAwAAAAAAUCE0ewAAAAAAACqEZg8AAAAAAECFmLv33MrMem5lLWKPPfYIjt9///3ROfr06ZNVOZ369Kc/HRx/+umnm14Dus7dLe8amoEs6nlHH310dJkLL7wwOD5s2LDg+EEHHRRdx9ixY4PjsaxCPsgiZGXLLbeMLjN//vzg+HrrrRcc33DDDbtUU54+8pGPBMcHDhwYnePJJ5/MqpwymOvuQ/MuImtkUdc88MAD0WVGjRoVHF+xYkXDdcT2z9mzZ0fn2HrrrRuuI+b555+PLrP77rsHx/v27Rudo8X+nqwriziyBwAAAAAAoEJo9gAAAAAAAFQIzR4AAAAAAIAKodkDAAAAAABQITR7AAAAAAAAKoRmDwAAAAAAQIXQ7AEAAAAAAKgQmj0AAAAAAAAV0jvvAtCY/fffPzjep0+fHqok7Omnn867BABNtP322wfHr7zyyugcI0aMCI7/5S9/6VJNHbnzzjsbngNAeZ1wwgnRZdw9OL777rtnVU7uXn311YbGgbLp3Tv+5+/pp58eHO/Vq1d0jlWrVgXHN9hgg+gcRxxxRHD8nHPOCY5vvfXW0XXEvP3229FlbrrppuD4aaedFp1j5cqVddeE+nFkDwAAAAAAQIXQ7AEAAAAAAKgQmj0AAAAAAAAVQrMHAAAAAACgQmj2AAAAAAAAVAjNHgAAAAAAgAqh2QMAAAAAAFAhvfMuAJ0bMmRIdJkzzjijByoJe+yxx6LL9O3bt+l1vPnmm9Fl3L3pdQCt6IgjjgiOP/jgg9E5Zs+eHRz/wAfC70/svffe0XVcdNFF0WUAlNduu+0WHD/rrLOic9x7773B8ccff7xLNQHoOZtuumlw/Lvf/W50juHDhwfH99xzz+gcI0eODI7Xk0U77rhjdJmQhQsXRpeZOnVqcPyOO+6IzjF//vy6a0LPih7ZY2ZbmtkvzWyBmT1pZqekj/czs/vN7E/p102aXy6AVkUWASgCsghAEZBFAGLqOY1rjaTT3H0HSbtLOsnMPiXpTEkz3f2Tkmam9wGgWcgiAEVAFgEoArIIQFC02ePuS9x9Xvr9G5IWSBoo6RBJU9LFpkg6tFlFAgBZBKAIyCIARUAWAYjp0jV7zGyQpJ0kPSKpv7svkZKwMbPNO3nOGEljGisTAN5DFgEoArIIQBGQRQA6Unezx8z6SrpD0nh3f93M6nqeu0+SNCmdg6vjAmgIWQSgCMgiAEVAFgHoTF0fvW5m6ykJkanufmf68FIzG5COD5C0rDklAkCCLAJQBGQRgCIgiwCE1PNpXCbpOkkL3P3SmqEZko5Jvz9G0vTsywOABFkEoAjIIgBFQBYBiKnnNK49JR0t6Qkzeyx97GxJF0m6zcyOl/RXSYc3p0QAkEQWASgGsghAEZBFAILMvedO0eR80Oz913/9V3B8xIgRPVRJWOz84Sx+D2+88cboMmvWrAmOT5o0KTrHnDlz6q6p7Ny9vhO/S4Ys6prBgwdHl3nssceC42eeGf/k1x/+8IfB8YEDBwbHZ8yYEV3HoYeGP5Tk2Wefjc6BnkcWoV7XX399cPxzn/tcdI4ddtghOB57LYFKm+vuQ/MuImtVyqKzzjorOH7hhRdG53jmmWeC46tXr47OMWvWrOD4HnvsEZ1j4403Do7/7ne/C45fffXV0XXcc8890WVQSHVlUV3X7AEAAAAAAEA50OwBAAAAAACoEJo9AAAAAAAAFUKzBwAAAAAAoEJo9gAAAAAAAFQIzR4AAAAAAIAKodkDAAAAAABQIebuPbcys55bWYvo379/cPxnP/tZdI6ddtopq3I6ZWbB8Z78PQxZtmxZdJlzzjknOH7DDTdkVU7u3D38gyspsqhrhgwZEl1m3rx5wfHNN988Osfy5cuD48cdd1xw/LOf/Wx0HePHj48ug+IhiyBJBx10UHSZO++8Mzg+ffr06ByHH3543TWh5cx196F5F5G1KmXRFltsERyfMWNGdI6hQ8M/4qlTp0bnOPHEE4PjK1eujM6x9dZbB8dXrVoVHH/ppZei60Bp1ZVFHNkDAAAAAABQITR7AAAAAAAAKoRmDwAAAAAAQIXQ7AEAAAAAAKgQmj0AAAAAAAAVQrMHAAAAAACgQmj2AAAAAAAAVAjNHgAAAAAAgArpnXcBaMzSpUuD4/vvv390jo985CMN1bDTTjtFl9l4442D49/+9rejc2yzzTZ119Rds2bNii7z4osvNr0OoGpGjx4dXebXv/51cPzkk08Ojv/0pz/tUk0AyuXcc8+NLvPSSy8Fx7/zne9kVQ6AAoq9Tt911117qJLGLV68OO8SUHIc2QMAAAAAAFAhNHsAAAAAAAAqhGYPAAAAAABAhdDsAQAAAAAAqBCaPQAAAAAAABVCswcAAAAAAKBCaPYAAAAAAABUiLl7z63MrOdWBqBh7m5519AMZFHXrLfeetFlrrzyyuD4/vvvH53j9ddfD44/8MADwfEzzjgjuo533303ugyKhyxqDV/4wheC4zNnzozOccsttwTHjzrqqC7VBLQz192H5l1E1sgioHTqyiKO7AEAAAAAAKgQmj0AAAAAAAAVQrMHAAAAAACgQmj2AAAAAAAAVAjNHgAAAAAAgAqh2QMAAAAAAFAhNHsAAAAAAAAqhGYPAAAAAABAhZi7hxcw21LSjZK2kLRW0iR3v9zMJkj6J0kvpYue7e4/j8wVXhmAQnF3y7uGNmRRsfXq1Ss4fs0110Tn2HjjjYPjo0ePDo6vWbMmug6UE1kEoCDmuvvQvIuQyCKgxdWVRb3rmGiNpNPcfZ6ZfUjSXDO7Px27zN0vaaRKAKgTWQSgCMgiAEVAFgEIijZ73H2JpCXp92+Y2QJJA5tdGADUIosAFAFZBKAIyCIAMV26Zo+ZDZK0k6RH0ofGmdnjZna9mW2ScW0A0CGyCEARkEUAioAsAtCRups9ZtZX0h2Sxrv765KukrStpCFKusoTO3neGDObY2ZzMqgXQIsjiwAUAVkEoAjIIgCdiV6gWZLMbD1JP5N0r7tf2sH4IEk/c/cdI/Nw8S+gRIp0UVSJLCoyLtCMZiKLABREYS7QLJFFQAurK4uiR/aYmUm6TtKC2hAxswE1i42UNL87VQJAPcgiAEVAFgEoArIIQEw9n8a1p6SjJT1hZo+lj50t6UgzGyLJJS2SNLYpFQJAgiwCUARkEYAiIIsABNV1GldmK+MQQaBUinbqRFbIIqBcyCIABVGo07iyQhYBpZPNaVwAAAAAAAAoD5o9AAAAAAAAFUKzBwAAAAAAoEJo9gAAAAAAAFQIzR4AAAAAAIAKodkDAAAAAABQITR7AAAAAAAAKoRmDwAAAAAAQIXQ7AEAAAAAAKgQmj0AAAAAAAAVQrMHAAAAAACgQmj2AAAAAAAAVAjNHgAAAAAAgAqh2QMAAAAAAFAhNHsAAAAAAAAqhGYPAAAAAABAhfTu4fUtl7S45v5m6WNFV5Y6pfLUSp3ZakadW2c8X5GUNYuk8tRKndlq5TrJomIqS63Uma1Wr7OqedQ+iyR+1lmjzmyVpU4px9dG5u4Zr7d+ZjbH3YfmVkCdylKnVJ5aqTNbZamzqMq0/cpSK3VmizpbQ5m2X1lqpc5sUWfrKMs2pM5sUWf28qyV07gAAAAAAAAqhGYPAAAAAABAheTd7JmU8/rrVZY6pfLUSp3ZKkudRVWm7VeWWqkzW9TZGsq0/cpSK3VmizpbR1m2IXVmizqzl1utuV6zBwAAAAAAANnK+8geAAAAAAAAZCi3Zo+ZDTezp81soZmdmVcdMWa2yMyeMLPHzGxO3vW0MbPrzWyZmc2veayfmd1vZn9Kv26SZ41tOql1gpk9n27Xx8zsgJxr3NLMfmlmC8zsSTM7JX28cNs0UGuhtmlZkEWNK0selSGL0ppKkUdkUbbIosaRRdkii1oTWdQ4sihbZFEDNeVxGpeZ9ZL0R0n7S3pO0qOSjnT3P/R4MRFmtkjSUHdfnncttcxsb0krJd3o7jumj10saYW7X5SG8ybu/q0860zr6qjWCZJWuvsledbWxswGSBrg7vPM7EOS5ko6VNKxKtg2DdR6hAq0TcuALMpGWfKoDFkklSePyKLskEXZIIuyRRa1HrIoG2RRtsii7svryJ5dJS1097+4+7uSbpF0SE61lJK7PyhpRbuHD5E0Jf1+ipJfrtx1UmuhuPsSd5+Xfv+GpAWSBqqA2zRQK7qOLMpAWfKoDFkklSePyKJMkUUZIIuyRRa1JLIoA2RRtsii7sur2TNQ0rM1959TcUPZJd1nZnPNbEzexUT0d/clUvLLJmnznOuJGWdmj6eHEOZ+KGMbMxskaSdJj6jg27RdrVJBt2mBkUXNU+h9p53C7jdlySOyqGFkUfMUdr/pQGH3G7KoZZBFzVPY/aYDhd1vyKKuyavZYx08VtSPBdvT3XeWNELSSenhbmjcVZK2lTRE0hJJE/MtJ2FmfSXdIWm8u7+edz0hHdRayG1acGQRCrvflCWPyKJMkEUo7H5DFrUUsgiF3W/Ioq7Lq9nznKQta+5/XNILOdUS5O4vpF+XSbpLyeGNRbU0PVew7ZzBZTnX0yl3X+ruf3P3tZKuUQG2q5mtp2THnOrud6YPF3KbdlRrEbdpCZBFzVPIfae9ou43ZckjsigzZFHzFG6/6UhR9xuyqOWQRc1TuP2mI0Xdb8ii7smr2fOopE+a2TZmtr6k0ZJm5FRLp8xso/TiSjKzjSR9UdL88LNyNUPSMen3x0ianmMtQW07Zmqkct6uZmaSrpO0wN0vrRkq3DbtrNaibdOSIIuap3D7TkeKuN+UJY/IokyRRc1TqP2mM0Xcb8iilkQWNU+h9pvOFHG/IYsaqMlz+DQuSbLkI8d+IKmXpOvd/cJcCgkws08o6RRLUm9JNxelTjP7iaRhkjaTtFTSeZJ+Kuk2SVtJ+qukw90994tudVLrMCWHsrmkRZLGtp1zmQcz20vSryU9IWlt+vDZSs6zLNQ2DdR6pAq0TcuCLGpcWfKoDFkklSePyKJskUWNI4uyRRa1JrKocWRRtsiiBmrKq9kDAAAAAACA7OV1GhcAAAAAAACagGYPAAAAAABAhdDsAQAAAAAAqBCaPQAAAAAAABVCswcAAAAAAKBCaPYAAAAAAABUCM0eAAAAAACACqHZAwAAAAAAUCH/H5Soz7+ZTu0sAAAAAElFTkSuQmCC\n",
       "text/plain": [
-       "<Figure size 1440x288 with 5 Axes>"
+       "<Figure size 1440x1440 with 16 Axes>"
       ]
      },
-     "metadata": {},
+     "metadata": {
+      "needs_background": "light"
+     },
      "output_type": "display_data"
     }
    ],
    "source": [
+    "\n",
     "import numpy as np \n",
+    "from random import sample\n",
     "import matplotlib.pyplot as plt\n",
-    "index = 0\n",
-    "misclassifiedIndexes = []\n",
-    "for label, predict in zip(y_test, predictions):\n",
-    " if label != predict: \n",
-    "  misclassifiedIndexes.append(index)\n",
-    "  index +=1\n",
+    "\n",
+    "def calculate_miclassified(x_test, y_test, clf):\n",
+    "    misclassified_inds = []\n",
+    "    predictions = clf.predict(x_test)\n",
+    "    for index, (label, predict) in enumerate(zip(y_test, predictions)):\n",
+    "        if label != predict: \n",
+    "            misclassified_inds.append(index)\n",
+    "    return misclassified_inds\n",
+    "\n",
+    "def plot_random_samples(x_test, y_test, predictions, misclassified_inds, cols = 4, rows = 4):\n",
+    "    num_images = cols * rows\n",
+    "    random_indexes = sample(misclassified_inds, num_images)\n",
+    "    fig=plt.figure(figsize=(20, 20))\n",
+    "    for i in range(1, cols*rows +1):\n",
+    "        img_index = random_indexes[i-1]\n",
+    "        img = np.reshape(x_test[img_index], (28,28))\n",
+    "        fig.add_subplot(rows, cols, i)\n",
+    "        plt.imshow(img, cmap=plt.cm.gray)\n",
+    "        plt.title('Predicted: {}, Actual: {}'.format(predictions[img_index], y_test[img_index]), fontsize = 15)\n",
+    "    plt.show()\n",
     "    \n",
-    "plt.figure(figsize=(20,4))\n",
-    "for plotIndex, badIndex in enumerate(misclassifiedIndexes[0:5]):\n",
-    " plt.subplot(1, 5, plotIndex + 1)\n",
-    " plt.imshow(np.reshape(x_test[badIndex], (28,28)), cmap=plt.cm.gray)\n",
-    " plt.title('Predicted: {}, Actual: {}'.format(predictions[badIndex], y_test[badIndex]), fontsize = 15)"
+    "misclassified_inds = calculate_miclassified(x_test, y_test, clf)\n",
+    "    \n",
+    "plot_random_samples(x_test, y_test, predictions, misclassified_inds)"
    ]
   },
   {
-   "cell_type": "code",
-   "execution_count": null,
+   "cell_type": "markdown",
    "metadata": {},
-   "outputs": [],
-   "source": []
+   "source": [
+    "### Conclusions\n",
+    "\n",
+    "So far we have built a classifier, that can predict hand written digits with ~88% accuracy. This is pretty good for starters, but it is just the beginning. Top hand written digit classifiers have accuracy almost 100% ! \n",
+    "\n",
+    "If you want to improve your classifier, you can follow this blogpost: https://towardsdatascience.com/a-simple-2d-cnn-for-mnist-digit-recognition-a998dbc1e79a"
+   ]
   },
   {
    "cell_type": "markdown",
@@ -528,12 +645,15 @@
    "source": [
     "### Implementing Logistic classification from scratch\n",
     "\n",
+    "The following section contains a simple implementation of the classifier built from scratch. \n",
+    "\n",
+    "Note: this is an advanced topic and you can skip if you are not interested in the underpinnings of the algorithm\n",
     "\n",
     "Algorithm\n",
     "Given a set of inputs X, we want to assign them to one of two possible categories (0 or 1). Logistic regression models the probability that each input belongs to a particular category.\n",
     "\n",
     "Hypothesis\n",
-    "A function takes inputs and returns outputs. To generate probabilities, logistic regression uses a function that gives outputs between 0 and 1 for all values of X. There are many functions that meet this description, but the used in this case is the logistic function. From here we will refer to it as softmax.\n",
+    "A function takes inputs and returns outputs. To generate probabilities, logistic regression uses a function that gives outputs between 0 and 1 for all values of X. There are many functions that meet this description, but the one used in this case is the logistic function. From here we will refer to it as softmax.\n",
     "n.\n",
     "\n"
    ]
@@ -551,19 +671,18 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 25,
+   "execution_count": 37,
    "metadata": {},
    "outputs": [],
    "source": [
     "\n",
     "def softmax(z):\n",
     "    \"\"\"\n",
-    "    Implement the softmax function\n",
+    "    The softmax function\n",
     "    Arguments:\n",
     "    z -- a k-length vector (float)\n",
     "    Return:\n",
-    "    result -- the softmax function evaluated on z, returning a set of probability \n",
-    "    of length k\n",
+    "    result -- the softmax function evaluated on z, returning a set of probability  of length k\n",
     "    \"\"\"\n",
     "    result = 1/sum(np.exp(z)) * np.exp(z)\n",
     "    return result\n",
@@ -581,12 +700,19 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 186,
+   "execution_count": 38,
    "metadata": {},
    "outputs": [],
    "source": [
     "\n",
     "def one_hot_encode(y):\n",
+    "    \"\"\"\n",
+    "    One hot encodings\n",
+    "    Arguments:\n",
+    "    y -- number between 0 and 9 that represents the digit\n",
+    "    Return:\n",
+    "    result -- 10 digit array, where all values but one is zero.\n",
+    "    \"\"\"\n",
     "    vector = np.array([0]*10)\n",
     "    vector[y] = 1\n",
     "    return vector\n"
@@ -596,7 +722,10 @@
    "cell_type": "markdown",
    "metadata": {},
    "source": [
-    "### Gradient"
+    "### Gradient Descent\n",
+    "\n",
+    "It is an optimization algorithm to find the minimum of a function. We start with a random point on the function and move in the negative direction of the gradient of the function to reach the local/global minima.\n",
+    "\n"
    ]
   },
   {
@@ -608,7 +737,7 @@
     "\n",
     "def gradient(w, x, y):\n",
     "    \"\"\"\n",
-    "    Implement the gradient \n",
+    "    The gradient \n",
     "    Arguments:\n",
     "    w -- weights, an ndarray of size (num_outputs, num_inputs)\n",
     "    x -- graphic data, a numpy array of size (num_inputs) -- the MNIST data representing pixels\n",
@@ -644,7 +773,7 @@
     "\n",
     "def initialize(num_outputs,num_inputs):\n",
     "    \"\"\"\n",
-    "    Initialize random weights w\n",
+    "    Initializing random weights w\n",
     "    \n",
     "    Return: \n",
     "    w -- weights, an ndarray of size (num_outputs, num_inputs)\n",
@@ -654,17 +783,14 @@
    ]
   },
   {
-   "cell_type": "code",
-   "execution_count": 176,
+   "cell_type": "markdown",
    "metadata": {},
-   "outputs": [],
    "source": [
+    "In machine learning, we can use a technique that evaluates and updates the coefficients every iteration called stochastic gradient descent to minimize the error of a model on our training data.\n",
+    "\n",
+    "The way this optimization algorithm works is that each training instance is shown to the model one at a time. The model makes a prediction for a training instance, the error is calculated and the model is updated in order to reduce the error for the next prediction.\n",
     "\n",
-    "def l_rate(base_rate, ite, num_iterations, schedule = False):\n",
-    "    if schedule == True:\n",
-    "        return base_rate * 10 ** (-np.floor(ite/num_iterations*4))\n",
-    "    else:\n",
-    "        return base_rate\n"
+    "This procedure can be used to find the set of coefficients in a model that result in the smallest error for the model on the training data."
    ]
   },
   {
@@ -676,14 +802,14 @@
     "\n",
     "def train(X_train, Y_train, num_iterations = 1000, learning_rate = 0.5):\n",
     "    \"\"\"\n",
-    "    Implement the stochastic gradient descent model \n",
+    "    The stochastic gradient descent model \n",
     "    Arguments:\n",
     "    X_train -- x_train data set, a 2darray of float in shape (num_training_data, num_inputs)\n",
     "    Y_train -- y_train data set, a vector of float in shape (num_training_data,)\n",
     "    num_iterations -- number of iterations to have\n",
     "    learning_rate -- size of base learning rate\n",
     "    Return:\n",
-    "    w -- the weights after optimization, an 2darray of size (num_outputs, num_inputs)\n",
+    "    w -- the weights after optimization, a 2darray of size (num_outputs, num_inputs)\n",
     "    \"\"\"\n",
     "    # initialize the random weights\n",
     "    w = initialize(num_outputs,num_inputs)\n",
@@ -769,7 +895,9 @@
   {
    "cell_type": "code",
    "execution_count": 244,
-   "metadata": {},
+   "metadata": {
+    "scrolled": true
+   },
    "outputs": [
     {
      "name": "stdout",
@@ -785,8 +913,6 @@
    ],
    "source": [
     "\n",
-    "\n",
-    "####################################################################################\n",
     "#Implementation of stochastic gradient descent algorithm\n",
     "#number of inputs\n",
     "num_inputs = 28*28\n",
@@ -806,6 +932,13 @@
     "        "
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "### Mnist classification using mini-batch approach(slow)"
+   ]
+  },
   {
    "cell_type": "markdown",
    "metadata": {},
@@ -846,7 +979,7 @@
     "\n",
     "def gradient_minibatch(W, X, Y):\n",
     "    \"\"\"\n",
-    "    Implement the gradient \n",
+    "    The gradient \n",
     "    Arguments:\n",
     "    w -- (num_outputs, num_inputs) weights, an ndarray of size (num_outputs, num_inputs)\n",
     "    x -- (mini_batch,  num_inputs) graphic data, a numpy array of size (num_inputs) -- the MNIST data representing pixels\n",
@@ -920,7 +1053,7 @@
     "\n",
     "def train_mini_batch(X_train, Y_train, num_iterations = 1000, learning_rate = 0.5, mini_batch = 64):\n",
     "    \"\"\"\n",
-    "    Implement the stochastic gradient descent model \n",
+    "    The stochastic gradient descent model \n",
     "    Arguments:\n",
     "    X_train -- x_train data set, a 2darray of float in shape (num_training_data, num_inputs)\n",
     "    Y_train -- y_train data set, a vector of float in shape (num_training_data,)\n",
@@ -979,9 +1112,6 @@
    ],
    "source": [
     "\n",
-    "\n",
-    "\n",
-    "####################################################################################\n",
     "#Implementation of stochastic gradient descent algorithm\n",
     "#number of inputs\n",
     "num_inputs = 28*28\n",
@@ -993,7 +1123,7 @@
     "\n",
     "w = train_mini_batch(x_train, y_train, num_iterations = num_iter, learning_rate = lr)\n",
     "print('Model #', (num_iter,lr))\n",
-    "print('Accuarcy on Test: ',testing(w, x_test, y_test))\n",
+    "print('Accuracy on Test: ',testing(w, x_test, y_test))\n",
     "        \n"
    ]
   },